feat: Add Hoodie comparison reports with scheduled jobs

- Migration for hoodie_comparison_reports table
- Comparison service: pulls all Hoodie data, compares to CannaIQ
- Stores delta results (hoodie_only, cannaiq_only, matched)
- Raw Hoodie data stays remote (proxy only)
- New endpoints:
  - POST /api/hoodie/reports/run/:state - run comparison
  - GET /api/hoodie/reports - latest reports
  - GET /api/hoodie/reports/:id - full report details
  - GET /api/hoodie/reports/history/:type/:state - report history

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

.gitignore

@@ -51,3 +51,10 @@ coverage/
 *.tmp
 *.temp
 llm-scraper/
+
+# Claude Code
+.claude/
+
+# Test/debug scripts
+backend/scripts/test-*.ts
+backend/scripts/run-*.ts

.woodpecker.yml

@@ -0,0 +1,189 @@
+steps:
+  # ===========================================
+  # PR VALIDATION: Parallel type checks (PRs only)
+  # ===========================================
+  typecheck-backend:
+    image: node:22
+    commands:
+      - cd backend
+      - npm ci --prefer-offline
+      - npx tsc --noEmit
+    depends_on: []
+    when:
+      event: pull_request
+
+  typecheck-cannaiq:
+    image: node:22
+    commands:
+      - cd cannaiq
+      - npm ci --prefer-offline
+      - npx tsc --noEmit
+    depends_on: []
+    when:
+      event: pull_request
+
+  typecheck-findadispo:
+    image: node:22
+    commands:
+      - cd findadispo/frontend
+      - npm ci --prefer-offline
+      - npx tsc --noEmit 2>/dev/null || true
+    depends_on: []
+    when:
+      event: pull_request
+
+  typecheck-findagram:
+    image: node:22
+    commands:
+      - cd findagram/frontend
+      - npm ci --prefer-offline
+      - npx tsc --noEmit 2>/dev/null || true
+    depends_on: []
+    when:
+      event: pull_request
+
+  # ===========================================
+  # AUTO-MERGE: Merge PR after all checks pass
+  # ===========================================
+  auto-merge:
+    image: alpine:latest
+    environment:
+      GITEA_TOKEN:
+        from_secret: gitea_token
+    commands:
+      - apk add --no-cache curl
+      - |
+        echo "Merging PR #${CI_COMMIT_PULL_REQUEST}..."
+        curl -s -X POST \
+          -H "Authorization: token $GITEA_TOKEN" \
+          -H "Content-Type: application/json" \
+          -d '{"Do":"merge"}' \
+          "https://git.spdy.io/api/v1/repos/Creationshop/cannaiq/pulls/${CI_COMMIT_PULL_REQUEST}/merge"
+    depends_on:
+      - typecheck-backend
+      - typecheck-cannaiq
+      - typecheck-findadispo
+      - typecheck-findagram
+    when:
+      event: pull_request
+
+  # ===========================================
+  # DOCKER: Multi-stage builds with layer caching
+  # ===========================================
+  docker-backend:
+    image: gcr.io/kaniko-project/executor:debug
+    commands:
+      - /kaniko/executor
+        --context=/woodpecker/src/git.spdy.io/Creationshop/cannaiq/backend
+        --dockerfile=/woodpecker/src/git.spdy.io/Creationshop/cannaiq/backend/Dockerfile
+        --destination=registry.spdy.io/cannaiq/backend:latest
+        --destination=registry.spdy.io/cannaiq/backend:sha-${CI_COMMIT_SHA:0:8}
+        --build-arg=APP_BUILD_VERSION=sha-${CI_COMMIT_SHA:0:8}
+        --build-arg=APP_GIT_SHA=${CI_COMMIT_SHA}
+        --build-arg=APP_BUILD_TIME=${CI_PIPELINE_CREATED}
+        --cache=true
+        --cache-repo=registry.spdy.io/cannaiq/cache-backend
+        --cache-ttl=168h
+    depends_on: []
+    when:
+      branch: [master, develop]
+      event: push
+
+  docker-cannaiq:
+    image: gcr.io/kaniko-project/executor:debug
+    commands:
+      - /kaniko/executor
+        --context=/woodpecker/src/git.spdy.io/Creationshop/cannaiq/cannaiq
+        --dockerfile=/woodpecker/src/git.spdy.io/Creationshop/cannaiq/cannaiq/Dockerfile
+        --destination=registry.spdy.io/cannaiq/frontend:latest
+        --destination=registry.spdy.io/cannaiq/frontend:sha-${CI_COMMIT_SHA:0:8}
+        --cache=true
+        --cache-repo=registry.spdy.io/cannaiq/cache-cannaiq
+        --cache-ttl=168h
+    depends_on: []
+    when:
+      branch: [master, develop]
+      event: push
+
+  docker-findadispo:
+    image: gcr.io/kaniko-project/executor:debug
+    commands:
+      - /kaniko/executor
+        --context=/woodpecker/src/git.spdy.io/Creationshop/cannaiq/findadispo/frontend
+        --dockerfile=/woodpecker/src/git.spdy.io/Creationshop/cannaiq/findadispo/frontend/Dockerfile
+        --destination=registry.spdy.io/cannaiq/findadispo:latest
+        --destination=registry.spdy.io/cannaiq/findadispo:sha-${CI_COMMIT_SHA:0:8}
+        --cache=true
+        --cache-repo=registry.spdy.io/cannaiq/cache-findadispo
+        --cache-ttl=168h
+    depends_on: []
+    when:
+      branch: [master, develop]
+      event: push
+
+  docker-findagram:
+    image: gcr.io/kaniko-project/executor:debug
+    commands:
+      - /kaniko/executor
+        --context=/woodpecker/src/git.spdy.io/Creationshop/cannaiq/findagram/frontend
+        --dockerfile=/woodpecker/src/git.spdy.io/Creationshop/cannaiq/findagram/frontend/Dockerfile
+        --destination=registry.spdy.io/cannaiq/findagram:latest
+        --destination=registry.spdy.io/cannaiq/findagram:sha-${CI_COMMIT_SHA:0:8}
+        --cache=true
+        --cache-repo=registry.spdy.io/cannaiq/cache-findagram
+        --cache-ttl=168h
+    depends_on: []
+    when:
+      branch: [master, develop]
+      event: push
+
+  # ===========================================
+  # DEPLOY: Pull from local registry
+  # ===========================================
+  deploy:
+    image: bitnami/kubectl:latest
+    environment:
+      K8S_TOKEN:
+        from_secret: k8s_token
+    commands:
+      - mkdir -p ~/.kube
+      - |
+        cat > ~/.kube/config << KUBEEOF
+        apiVersion: v1
+        kind: Config
+        clusters:
+        - cluster:
+            certificate-authority-data: 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
+            server: https://10.100.6.10:6443
+          name: spdy-k3s
+        contexts:
+        - context:
+            cluster: spdy-k3s
+            namespace: cannaiq
+            user: cannaiq-admin
+          name: cannaiq
+        current-context: cannaiq
+        users:
+        - name: cannaiq-admin
+          user:
+            token: $K8S_TOKEN
+        KUBEEOF
+      - chmod 600 ~/.kube/config
+      # Apply manifests to ensure probes and resource limits are set
+      - kubectl apply -f /woodpecker/src/git.spdy.io/Creationshop/cannaiq/k8s/scraper.yaml
+      - kubectl apply -f /woodpecker/src/git.spdy.io/Creationshop/cannaiq/k8s/scraper-worker.yaml
+      - kubectl set image deployment/scraper scraper=registry.spdy.io/cannaiq/backend:sha-${CI_COMMIT_SHA:0:8} -n cannaiq
+      - kubectl rollout status deployment/scraper -n cannaiq --timeout=300s
+      - kubectl set image deployment/scraper-worker worker=registry.spdy.io/cannaiq/backend:sha-${CI_COMMIT_SHA:0:8} -n cannaiq
+      - kubectl set image deployment/cannaiq-frontend cannaiq-frontend=registry.spdy.io/cannaiq/frontend:sha-${CI_COMMIT_SHA:0:8} -n cannaiq
+      - kubectl set image deployment/findadispo-frontend findadispo-frontend=registry.spdy.io/cannaiq/findadispo:sha-${CI_COMMIT_SHA:0:8} -n cannaiq
+      - kubectl set image deployment/findagram-frontend findagram-frontend=registry.spdy.io/cannaiq/findagram:sha-${CI_COMMIT_SHA:0:8} -n cannaiq
+      - kubectl rollout status deployment/cannaiq-frontend -n cannaiq --timeout=300s
+    depends_on:
+      - docker-backend
+      - docker-cannaiq
+      - docker-findadispo
+      - docker-findagram
+    when:
+      branch: [master, develop]
+      event: push

.woodpecker/.ci.yml

@@ -1,140 +0,0 @@
-when:
-  - event: [push, pull_request]
-
-steps:
-  # Build checks
-  typecheck-backend:
-    image: node:20
-    commands:
-      - cd backend
-      - npm ci
-      - npx tsc --noEmit || true
-
-  build-cannaiq:
-    image: node:20
-    commands:
-      - cd cannaiq
-      - npm ci
-      - npx tsc --noEmit
-      - npm run build
-
-  build-findadispo:
-    image: node:20
-    commands:
-      - cd findadispo/frontend
-      - npm ci
-      - npm run build
-
-  build-findagram:
-    image: node:20
-    commands:
-      - cd findagram/frontend
-      - npm ci
-      - npm run build
-
-  # Docker builds - only on master
-  docker-backend:
-    image: woodpeckerci/plugin-docker-buildx
-    settings:
-      registry: code.cannabrands.app
-      repo: code.cannabrands.app/creationshop/dispensary-scraper
-      tags:
-        - latest
-        - ${CI_COMMIT_SHA:0:8}
-      dockerfile: backend/Dockerfile
-      context: backend
-      username:
-        from_secret: registry_username
-      password:
-        from_secret: registry_password
-      platforms: linux/amd64
-      provenance: false
-    when:
-      branch: master
-      event: push
-
-  docker-cannaiq:
-    image: woodpeckerci/plugin-docker-buildx
-    settings:
-      registry: code.cannabrands.app
-      repo: code.cannabrands.app/creationshop/cannaiq-frontend
-      tags:
-        - latest
-        - ${CI_COMMIT_SHA:0:8}
-      dockerfile: cannaiq/Dockerfile
-      context: cannaiq
-      username:
-        from_secret: registry_username
-      password:
-        from_secret: registry_password
-      platforms: linux/amd64
-      provenance: false
-    when:
-      branch: master
-      event: push
-
-  docker-findadispo:
-    image: woodpeckerci/plugin-docker-buildx
-    settings:
-      registry: code.cannabrands.app
-      repo: code.cannabrands.app/creationshop/findadispo-frontend
-      tags:
-        - latest
-        - ${CI_COMMIT_SHA:0:8}
-      dockerfile: findadispo/frontend/Dockerfile
-      context: findadispo/frontend
-      username:
-        from_secret: registry_username
-      password:
-        from_secret: registry_password
-      platforms: linux/amd64
-      provenance: false
-    when:
-      branch: master
-      event: push
-
-  docker-findagram:
-    image: woodpeckerci/plugin-docker-buildx
-    settings:
-      registry: code.cannabrands.app
-      repo: code.cannabrands.app/creationshop/findagram-frontend
-      tags:
-        - latest
-        - ${CI_COMMIT_SHA:0:8}
-      dockerfile: findagram/frontend/Dockerfile
-      context: findagram/frontend
-      username:
-        from_secret: registry_username
-      password:
-        from_secret: registry_password
-      platforms: linux/amd64
-      provenance: false
-    when:
-      branch: master
-      event: push
-
-  # Deploy to Kubernetes
-  deploy:
-    image: bitnami/kubectl:latest
-    environment:
-      KUBECONFIG_CONTENT:
-        from_secret: kubeconfig_data
-    commands:
-      - echo "Deploying to Kubernetes..."
-      - mkdir -p ~/.kube
-      - echo "$KUBECONFIG_CONTENT" | tr -d '[:space:]' | base64 -d > ~/.kube/config
-      - chmod 600 ~/.kube/config
-      - kubectl set image deployment/scraper scraper=code.cannabrands.app/creationshop/dispensary-scraper:${CI_COMMIT_SHA:0:8} -n dispensary-scraper
-      - kubectl set image deployment/scraper-worker scraper-worker=code.cannabrands.app/creationshop/dispensary-scraper:${CI_COMMIT_SHA:0:8} -n dispensary-scraper
-      - kubectl set image deployment/cannaiq-frontend cannaiq-frontend=code.cannabrands.app/creationshop/cannaiq-frontend:${CI_COMMIT_SHA:0:8} -n dispensary-scraper
-      - kubectl set image deployment/findadispo-frontend findadispo-frontend=code.cannabrands.app/creationshop/findadispo-frontend:${CI_COMMIT_SHA:0:8} -n dispensary-scraper
-      - kubectl set image deployment/findagram-frontend findagram-frontend=code.cannabrands.app/creationshop/findagram-frontend:${CI_COMMIT_SHA:0:8} -n dispensary-scraper
-      - kubectl rollout status deployment/scraper -n dispensary-scraper --timeout=300s
-      - kubectl rollout status deployment/scraper-worker -n dispensary-scraper --timeout=300s
-      - kubectl rollout status deployment/cannaiq-frontend -n dispensary-scraper --timeout=120s
-      - kubectl rollout status deployment/findadispo-frontend -n dispensary-scraper --timeout=120s
-      - kubectl rollout status deployment/findagram-frontend -n dispensary-scraper --timeout=120s
-      - echo "All deployments complete!"
-    when:
-      branch: master
-      event: push

backend/.env

@@ -1,30 +1,52 @@
+# CannaiQ Backend Environment Configuration
+# Copy this file to .env and fill in the values
+
+# Server
 PORT=3010
 NODE_ENV=development
 
 # =============================================================================
-# CannaiQ Database (dutchie_menus) - PRIMARY DATABASE
+# CANNAIQ DATABASE (dutchie_menus) - PRIMARY DATABASE
 # =============================================================================
-# This is where all schema migrations run and where canonical tables live.
+# This is where ALL schema migrations run and where canonical tables live.
-# All CANNAIQ_DB_* variables are REQUIRED - connection will fail if missing.
+# All CANNAIQ_DB_* variables are REQUIRED - no defaults.
+# The application will fail to start if any are missing.
+
 CANNAIQ_DB_HOST=localhost
 CANNAIQ_DB_PORT=54320
-CANNAIQ_DB_NAME=dutchie_menus
+CANNAIQ_DB_NAME=dutchie_menus   # MUST be dutchie_menus - NOT dutchie_legacy
 CANNAIQ_DB_USER=dutchie
 CANNAIQ_DB_PASS=dutchie_local_pass
 
+# Alternative: Use a full connection URL instead of individual vars
+# If set, this takes priority over individual vars above
+# CANNAIQ_DB_URL=postgresql://user:pass@host:port/dutchie_menus
+
 # =============================================================================
-# Legacy Database (dutchie_legacy) - READ-ONLY SOURCE
+# LEGACY DATABASE (dutchie_legacy) - READ-ONLY FOR ETL
 # =============================================================================
 # Used ONLY by ETL scripts to read historical data.
 # NEVER run migrations against this database.
+# These are only needed when running 042_legacy_import.ts
+
 LEGACY_DB_HOST=localhost
 LEGACY_DB_PORT=54320
-LEGACY_DB_NAME=dutchie_legacy
+LEGACY_DB_NAME=dutchie_legacy   # READ-ONLY - never migrated
 LEGACY_DB_USER=dutchie
-LEGACY_DB_PASS=dutchie_local_pass
+LEGACY_DB_PASS=
 
-# Local image storage (no MinIO per CLAUDE.md)
+# Alternative: Use a full connection URL instead of individual vars
+# LEGACY_DB_URL=postgresql://user:pass@host:port/dutchie_legacy
+
+# =============================================================================
+# LOCAL STORAGE
+# =============================================================================
+# Local image storage path (no MinIO)
 LOCAL_IMAGES_PATH=./public/images
 
-# JWT
+# =============================================================================
+# AUTHENTICATION
+# =============================================================================
 JWT_SECRET=your-secret-key-change-in-production
+ANTHROPIC_API_KEY=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
+OPENAI_API_KEY=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

backend/.gitignore

@@ -0,0 +1,3 @@
+
+# IP2Location database (downloaded separately)
+data/ip2location/

backend/Dockerfile

@@ -1,17 +1,33 @@
 # Build stage
-# Image: code.cannabrands.app/creationshop/dispensary-scraper
+# Image: git.spdy.io/creationshop/dispensary-scraper
-FROM node:20-slim AS builder
+FROM node:22-slim AS builder
+
+# Install build tools for native modules (bcrypt, sharp)
+RUN apt-get update && apt-get install -y \
+    python3 \
+    build-essential \
+    --no-install-recommends \
+    && rm -rf /var/lib/apt/lists/*
 
 WORKDIR /app
 
 COPY package*.json ./
-RUN npm ci
+
+# Install dependencies with retry and fallback registry
+RUN npm config set fetch-retries 3 && \
+    npm config set fetch-retry-mintimeout 20000 && \
+    npm config set fetch-retry-maxtimeout 120000 && \
+    npm install || \
+    (npm config set registry https://registry.npmmirror.com && npm install)
 
 COPY . .
 RUN npm run build
 
+# Prune dev dependencies for smaller production image
+RUN npm prune --production
+
 # Production stage
-FROM node:20-slim
+FROM node:22-slim
 
 # Build arguments for version info
 ARG APP_BUILD_VERSION=dev
@@ -25,8 +41,10 @@ ENV APP_GIT_SHA=${APP_GIT_SHA}
 ENV APP_BUILD_TIME=${APP_BUILD_TIME}
 ENV CONTAINER_IMAGE_TAG=${CONTAINER_IMAGE_TAG}
 
-# Install Chromium dependencies
+# Install Chromium dependencies, curl, and CA certificates for HTTPS
 RUN apt-get update && apt-get install -y \
+    curl \
+    ca-certificates \
     chromium \
     fonts-liberation \
     libnss3 \
@@ -43,10 +61,12 @@ ENV PUPPETEER_EXECUTABLE_PATH=/usr/bin/chromium
 WORKDIR /app
 
 COPY package*.json ./
-RUN npm ci --omit=dev
+COPY --from=builder /app/node_modules ./node_modules
-
 COPY --from=builder /app/dist ./dist
 
+# Copy migrations for auto-migrate on startup
+COPY migrations ./migrations
+
 # Create local images directory for when MinIO is not configured
 RUN mkdir -p /app/public/images/products
 

backend/docs/CODEBASE_MAP.md

@@ -0,0 +1,268 @@
+# CannaiQ Backend Codebase Map
+
+**Last Updated:** 2025-12-12
+**Purpose:** Help Claude and developers understand which code is current vs deprecated
+
+---
+
+## Quick Reference: What to Use
+
+### For Crawling/Scraping
+| Task | Use This | NOT This |
+|------|----------|----------|
+| Fetch products | `src/tasks/handlers/payload-fetch.ts` | `src/hydration/*` |
+| Process products | `src/tasks/handlers/product-refresh.ts` | `src/scraper-v2/*` |
+| GraphQL client | `src/platforms/dutchie/client.ts` | `src/dutchie-az/services/graphql-client.ts` |
+| Worker system | `src/tasks/task-worker.ts` | `src/dutchie-az/services/worker.ts` |
+
+### For Database
+| Task | Use This | NOT This |
+|------|----------|----------|
+| Get DB pool | `src/db/pool.ts` | `src/dutchie-az/db/connection.ts` |
+| Run migrations | `src/db/migrate.ts` (CLI only) | Never import at runtime |
+| Query products | `store_products` table | `products`, `dutchie_products` |
+| Query stores | `dispensaries` table | `stores` table |
+
+### For Discovery
+| Task | Use This |
+|------|----------|
+| Discover stores | `src/discovery/*.ts` |
+| Run discovery | `npx tsx src/scripts/run-discovery.ts` |
+
+---
+
+## Directory Status
+
+### ACTIVE DIRECTORIES (Use These)
+
+```
+src/
+├── auth/               # JWT/session auth, middleware
+├── db/                 # Database pool, migrations
+├── discovery/          # Dutchie store discovery pipeline
+├── middleware/         # Express middleware
+├── multi-state/        # Multi-state query support
+├── platforms/          # Platform-specific clients (Dutchie, Jane, etc)
+│   └── dutchie/        # THE Dutchie client - use this one
+├── routes/             # Express API routes
+├── services/           # Core services (logger, scheduler, etc)
+├── tasks/              # Task system (workers, handlers, scheduler)
+│   └── handlers/       # Task handlers (payload_fetch, product_refresh, etc)
+├── types/              # TypeScript types
+└── utils/              # Utilities (storage, image processing)
+```
+
+### DEPRECATED DIRECTORIES (DO NOT USE)
+
+```
+src/
+├── hydration/          # DEPRECATED - Old pipeline approach
+├── scraper-v2/         # DEPRECATED - Old scraper engine
+├── canonical-hydration/# DEPRECATED - Merged into tasks/handlers
+├── dutchie-az/         # PARTIAL - Some parts deprecated, some active
+│   ├── db/             # DEPRECATED - Use src/db/pool.ts
+│   └── services/       # PARTIAL - worker.ts still runs, graphql-client.ts deprecated
+├── portals/            # FUTURE - Not yet implemented
+├── seo/                # PARTIAL - Settings work, templates WIP
+└── system/             # DEPRECATED - Old orchestration system
+```
+
+### DEPRECATED FILES (DO NOT USE)
+
+```
+src/dutchie-az/db/connection.ts      # Use src/db/pool.ts instead
+src/dutchie-az/services/graphql-client.ts  # Use src/platforms/dutchie/client.ts
+src/hydration/*.ts                   # Entire directory deprecated
+src/scraper-v2/*.ts                  # Entire directory deprecated
+```
+
+---
+
+## Key Files Reference
+
+### Entry Points
+| File | Purpose | Status |
+|------|---------|--------|
+| `src/index.ts` | Main Express server | ACTIVE |
+| `src/dutchie-az/services/worker.ts` | Worker process entry | ACTIVE |
+| `src/tasks/task-worker.ts` | Task worker (new system) | ACTIVE |
+
+### Dutchie Integration
+| File | Purpose | Status |
+|------|---------|--------|
+| `src/platforms/dutchie/client.ts` | GraphQL client, hashes, curl | **PRIMARY** |
+| `src/platforms/dutchie/queries.ts` | High-level query functions | ACTIVE |
+| `src/platforms/dutchie/index.ts` | Re-exports | ACTIVE |
+
+### Task Handlers
+| File | Purpose | Status |
+|------|---------|--------|
+| `src/tasks/handlers/payload-fetch.ts` | Fetch products from Dutchie | **PRIMARY** |
+| `src/tasks/handlers/product-refresh.ts` | Process payload into DB | **PRIMARY** |
+| `src/tasks/handlers/entry-point-discovery.ts` | Resolve platform IDs (auto-healing) | **PRIMARY** |
+| `src/tasks/handlers/menu-detection.ts` | Detect menu type | ACTIVE |
+| `src/tasks/handlers/id-resolution.ts` | Resolve platform IDs (legacy) | LEGACY |
+| `src/tasks/handlers/image-download.ts` | Download product images | ACTIVE |
+
+---
+
+## Transport Rules (CRITICAL)
+
+**Browser-based (Puppeteer) is the DEFAULT transport. curl is ONLY allowed when explicitly specified.**
+
+### Transport Selection
+| `task.method` | Transport Used | Notes |
+|---------------|----------------|-------|
+| `null` | Browser (Puppeteer) | DEFAULT - use this for most tasks |
+| `'http'` | Browser (Puppeteer) | Explicit browser request |
+| `'curl'` | curl-impersonate | ONLY when explicitly needed |
+
+### Why Browser-First?
+1. **Anti-detection**: Puppeteer with StealthPlugin evades bot detection
+2. **Session cookies**: Browser maintains session state automatically
+3. **Fingerprinting**: Real browser fingerprint (TLS, headers, etc.)
+4. **Age gates**: Browser can click through age verification
+
+### Entry Point Discovery Auto-Healing
+The `entry_point_discovery` handler uses a healing strategy:
+
+```
+1. FIRST: Check dutchie_discovery_locations for existing platform_location_id
+   - By linked dutchie_discovery_id
+   - By slug match in discovery data
+   → If found, NO network call needed
+
+2. SECOND: Browser-based GraphQL (Puppeteer)
+   - 5x retries for network/proxy failures
+   - On HTTP 403: rotate proxy and retry
+   - On HTTP 404 after 2 attempts: mark as 'removed'
+
+3. HARD FAILURE: After exhausting options → 'needs_investigation'
+```
+
+### DO NOT Use curl Unless:
+- Task explicitly has `method = 'curl'`
+- You're testing curl-impersonate binaries
+- The API explicitly requires curl fingerprinting
+
+### Files
+| File | Transport | Purpose |
+|------|-----------|---------|
+| `src/services/puppeteer-preflight.ts` | Browser | Preflight check |
+| `src/services/curl-preflight.ts` | curl | Preflight check |
+| `src/tasks/handlers/entry-point-discovery.ts` | Browser | Platform ID resolution |
+| `src/tasks/handlers/payload-fetch.ts` | Both | Product fetching |
+
+### Database
+| File | Purpose | Status |
+|------|---------|--------|
+| `src/db/pool.ts` | Canonical DB pool | **PRIMARY** |
+| `src/db/migrate.ts` | Migration runner (CLI only) | CLI ONLY |
+| `src/db/auto-migrate.ts` | Auto-run migrations on startup | ACTIVE |
+
+### Configuration
+| File | Purpose | Status |
+|------|---------|--------|
+| `.env` | Environment variables | ACTIVE |
+| `package.json` | Dependencies | ACTIVE |
+| `tsconfig.json` | TypeScript config | ACTIVE |
+
+---
+
+## GraphQL Hashes (CRITICAL)
+
+The correct hashes are in `src/platforms/dutchie/client.ts`:
+
+```typescript
+export const GRAPHQL_HASHES = {
+  FilteredProducts: 'ee29c060826dc41c527e470e9ae502c9b2c169720faa0a9f5d25e1b9a530a4a0',
+  GetAddressBasedDispensaryData: '13461f73abf7268770dfd05fe7e10c523084b2bb916a929c08efe3d87531977b',
+  ConsumerDispensaries: '0a5bfa6ca1d64ae47bcccb7c8077c87147cbc4e6982c17ceec97a2a4948b311b',
+  GetAllCitiesByState: 'ae547a0466ace5a48f91e55bf6699eacd87e3a42841560f0c0eabed5a0a920e6',
+};
+```
+
+**ALWAYS** use `Status: 'Active'` for FilteredProducts (not `null` or `'All'`).
+
+---
+
+## Scripts Reference
+
+### Useful Scripts (in `src/scripts/`)
+| Script | Purpose |
+|--------|---------|
+| `run-discovery.ts` | Run Dutchie discovery |
+| `crawl-single-store.ts` | Test crawl a single store |
+| `test-dutchie-graphql.ts` | Test GraphQL queries |
+
+### One-Off Scripts (probably don't need)
+| Script | Purpose |
+|--------|---------|
+| `harmonize-az-dispensaries.ts` | One-time data cleanup |
+| `bootstrap-stores-for-dispensaries.ts` | One-time migration |
+| `backfill-*.ts` | Historical backfill scripts |
+
+---
+
+## API Routes
+
+### Active Routes (in `src/routes/`)
+| Route File | Mount Point | Purpose |
+|------------|-------------|---------|
+| `auth.ts` | `/api/auth` | Login/logout/session |
+| `stores.ts` | `/api/stores` | Store CRUD |
+| `dashboard.ts` | `/api/dashboard` | Dashboard stats |
+| `workers.ts` | `/api/workers` | Worker monitoring |
+| `pipeline.ts` | `/api/pipeline` | Crawl triggers |
+| `discovery.ts` | `/api/discovery` | Discovery management |
+| `analytics.ts` | `/api/analytics` | Analytics queries |
+| `wordpress.ts` | `/api/v1/wordpress` | WordPress plugin API |
+
+---
+
+## Documentation Files
+
+### Current Docs (in `backend/docs/`)
+| Doc | Purpose | Currency |
+|-----|---------|----------|
+| `TASK_WORKFLOW_2024-12-10.md` | Task system architecture | CURRENT |
+| `WORKER_TASK_ARCHITECTURE.md` | Worker/task design | CURRENT |
+| `CRAWL_PIPELINE.md` | Crawl pipeline overview | CURRENT |
+| `ORGANIC_SCRAPING_GUIDE.md` | Browser-based scraping | CURRENT |
+| `CODEBASE_MAP.md` | This file | CURRENT |
+| `ANALYTICS_V2_EXAMPLES.md` | Analytics API examples | CURRENT |
+| `BRAND_INTELLIGENCE_API.md` | Brand API docs | CURRENT |
+
+### Root Docs
+| Doc | Purpose | Currency |
+|-----|---------|----------|
+| `CLAUDE.md` | Claude instructions | **PRIMARY** |
+| `README.md` | Project overview | NEEDS UPDATE |
+
+---
+
+## Common Mistakes to Avoid
+
+1. **Don't use `src/hydration/`** - It's an old approach that was superseded by the task system
+
+2. **Don't use `src/dutchie-az/db/connection.ts`** - Use `src/db/pool.ts` instead
+
+3. **Don't import `src/db/migrate.ts` at runtime** - It will crash. Only use for CLI migrations.
+
+4. **Don't query `stores` table** - It's empty. Use `dispensaries`.
+
+5. **Don't query `products` table** - It's empty. Use `store_products`.
+
+6. **Don't use wrong GraphQL hash** - Always get hash from `GRAPHQL_HASHES` in client.ts
+
+7. **Don't use `Status: null`** - It returns 0 products. Use `Status: 'Active'`.
+
+---
+
+## When in Doubt
+
+1. Check if the file is imported in `src/index.ts` - if not, it may be deprecated
+2. Check the last modified date - older files may be stale
+3. Look for `DEPRECATED` comments in the code
+4. Ask: "Is there a newer version of this in `src/tasks/` or `src/platforms/`?"
+5. Read the relevant doc in `docs/` before modifying code

backend/docs/QUERY_API.md

@@ -0,0 +1,343 @@
+# CannaiQ Query API
+
+Query raw crawl payload data with flexible filters, sorting, and aggregation.
+
+## Base URL
+
+```
+https://cannaiq.co/api/payloads
+```
+
+## Authentication
+
+Include your API key in the header:
+```
+X-API-Key: your-api-key
+```
+
+---
+
+## Endpoints
+
+### 1. Query Products
+
+Filter and search products from a store's latest crawl data.
+
+```
+GET /api/payloads/store/{dispensaryId}/query
+```
+
+#### Query Parameters
+
+| Parameter | Type | Description |
+|-----------|------|-------------|
+| `brand` | string | Filter by brand name (partial match) |
+| `category` | string | Filter by category (flower, vape, edible, etc.) |
+| `subcategory` | string | Filter by subcategory |
+| `strain_type` | string | Filter by strain (indica, sativa, hybrid, cbd) |
+| `in_stock` | boolean | Filter by stock status (true/false) |
+| `price_min` | number | Minimum price |
+| `price_max` | number | Maximum price |
+| `thc_min` | number | Minimum THC percentage |
+| `thc_max` | number | Maximum THC percentage |
+| `search` | string | Search product name (partial match) |
+| `fields` | string | Comma-separated fields to return |
+| `limit` | number | Max results (default 100, max 1000) |
+| `offset` | number | Skip results for pagination |
+| `sort` | string | Sort by: name, price, thc, brand |
+| `order` | string | Sort order: asc, desc |
+
+#### Available Fields
+
+When using `fields` parameter, you can request:
+- `id` - Product ID
+- `name` - Product name
+- `brand` - Brand name
+- `category` - Product category
+- `subcategory` - Product subcategory
+- `strain_type` - Indica/Sativa/Hybrid/CBD
+- `price` - Current price
+- `price_med` - Medical price
+- `price_rec` - Recreational price
+- `thc` - THC percentage
+- `cbd` - CBD percentage
+- `weight` - Product weight/size
+- `status` - Stock status
+- `in_stock` - Boolean in-stock flag
+- `image_url` - Product image
+- `description` - Product description
+
+#### Examples
+
+**Get all flower products under $40:**
+```
+GET /api/payloads/store/112/query?category=flower&price_max=40
+```
+
+**Search for "Blue Dream" with high THC:**
+```
+GET /api/payloads/store/112/query?search=blue+dream&thc_min=20
+```
+
+**Get only name and price for Alien Labs products:**
+```
+GET /api/payloads/store/112/query?brand=Alien+Labs&fields=name,price,thc
+```
+
+**Get top 10 highest THC products:**
+```
+GET /api/payloads/store/112/query?sort=thc&order=desc&limit=10
+```
+
+**Paginate through in-stock products:**
+```
+GET /api/payloads/store/112/query?in_stock=true&limit=50&offset=0
+GET /api/payloads/store/112/query?in_stock=true&limit=50&offset=50
+```
+
+#### Response
+
+```json
+{
+  "success": true,
+  "dispensaryId": 112,
+  "payloadId": 45,
+  "fetchedAt": "2025-12-11T10:30:00Z",
+  "query": {
+    "filters": {
+      "brand": "Alien Labs",
+      "category": null,
+      "price_max": null
+    },
+    "sort": "price",
+    "order": "asc",
+    "limit": 100,
+    "offset": 0
+  },
+  "pagination": {
+    "total": 15,
+    "returned": 15,
+    "limit": 100,
+    "offset": 0,
+    "has_more": false
+  },
+  "products": [
+    {
+      "id": "507f1f77bcf86cd799439011",
+      "name": "Alien Labs - Baklava 3.5g",
+      "brand": "Alien Labs",
+      "category": "flower",
+      "strain_type": "hybrid",
+      "price": 55,
+      "thc": "28.5",
+      "in_stock": true
+    }
+  ]
+}
+```
+
+---
+
+### 2. Aggregate Data
+
+Group products and calculate metrics.
+
+```
+GET /api/payloads/store/{dispensaryId}/aggregate
+```
+
+#### Query Parameters
+
+| Parameter | Type | Description |
+|-----------|------|-------------|
+| `group_by` | string | **Required.** Field to group by: brand, category, subcategory, strain_type |
+| `metrics` | string | Comma-separated metrics (default: count) |
+
+#### Available Metrics
+
+- `count` - Number of products
+- `avg_price` - Average price
+- `min_price` - Lowest price
+- `max_price` - Highest price
+- `avg_thc` - Average THC percentage
+- `in_stock_count` - Number of in-stock products
+
+#### Examples
+
+**Count products by brand:**
+```
+GET /api/payloads/store/112/aggregate?group_by=brand
+```
+
+**Get price stats by category:**
+```
+GET /api/payloads/store/112/aggregate?group_by=category&metrics=count,avg_price,min_price,max_price
+```
+
+**Get THC averages by strain type:**
+```
+GET /api/payloads/store/112/aggregate?group_by=strain_type&metrics=count,avg_thc
+```
+
+**Brand analysis with stock info:**
+```
+GET /api/payloads/store/112/aggregate?group_by=brand&metrics=count,avg_price,in_stock_count
+```
+
+#### Response
+
+```json
+{
+  "success": true,
+  "dispensaryId": 112,
+  "payloadId": 45,
+  "fetchedAt": "2025-12-11T10:30:00Z",
+  "groupBy": "brand",
+  "metrics": ["count", "avg_price"],
+  "totalProducts": 450,
+  "groupCount": 85,
+  "aggregations": [
+    {
+      "brand": "Alien Labs",
+      "count": 15,
+      "avg_price": 52.33
+    },
+    {
+      "brand": "Connected",
+      "count": 12,
+      "avg_price": 48.50
+    }
+  ]
+}
+```
+
+---
+
+### 3. Compare Stores (Price Comparison)
+
+Query the same data from multiple stores and compare in your app:
+
+```javascript
+// Get flower prices from Store A
+const storeA = await fetch('/api/payloads/store/112/query?category=flower&fields=name,brand,price');
+
+// Get flower prices from Store B
+const storeB = await fetch('/api/payloads/store/115/query?category=flower&fields=name,brand,price');
+
+// Compare in your app
+const dataA = await storeA.json();
+const dataB = await storeB.json();
+
+// Find matching products and compare prices
+```
+
+---
+
+### 4. Price History
+
+For historical price data, use the snapshots endpoint:
+
+```
+GET /api/v1/products/{productId}/history?days=30
+```
+
+Or compare payloads over time:
+
+```
+GET /api/payloads/store/{dispensaryId}/diff?from={payloadId1}&to={payloadId2}
+```
+
+The diff endpoint shows:
+- Products added
+- Products removed
+- Price changes
+- Stock changes
+
+---
+
+### 5. List Stores
+
+Get available dispensaries to query:
+
+```
+GET /api/stores
+```
+
+Returns all stores with their IDs, names, and locations.
+
+---
+
+## Use Cases
+
+### Price Comparison App
+
+```javascript
+// 1. Get stores in Arizona
+const stores = await fetch('/api/stores?state=AZ').then(r => r.json());
+
+// 2. Query flower prices from each store
+const prices = await Promise.all(
+  stores.map(store =>
+    fetch(`/api/payloads/store/${store.id}/query?category=flower&fields=name,brand,price`)
+      .then(r => r.json())
+  )
+);
+
+// 3. Build comparison matrix in your app
+```
+
+### Brand Analytics Dashboard
+
+```javascript
+// Get brand presence across stores
+const brandData = await Promise.all(
+  storeIds.map(id =>
+    fetch(`/api/payloads/store/${id}/aggregate?group_by=brand&metrics=count,avg_price`)
+      .then(r => r.json())
+  )
+);
+
+// Aggregate brand presence across all stores
+```
+
+### Deal Finder
+
+```javascript
+// Find high-THC flower under $30
+const deals = await fetch(
+  '/api/payloads/store/112/query?category=flower&price_max=30&thc_min=20&in_stock=true&sort=thc&order=desc'
+).then(r => r.json());
+```
+
+### Inventory Tracker
+
+```javascript
+// Get products that went out of stock
+const diff = await fetch('/api/payloads/store/112/diff').then(r => r.json());
+
+const outOfStock = diff.details.stockChanges.filter(
+  p => p.newStatus !== 'Active'
+);
+```
+
+---
+
+## Rate Limits
+
+- Default: 100 requests/minute per API key
+- Contact support for higher limits
+
+## Error Responses
+
+```json
+{
+  "success": false,
+  "error": "Error message here"
+}
+```
+
+Common errors:
+- `404` - Store or payload not found
+- `400` - Missing required parameter
+- `401` - Invalid or missing API key
+- `429` - Rate limit exceeded

backend/docs/_archive/BRAND_INTELLIGENCE_API.md

@@ -0,0 +1,394 @@
+# Brand Intelligence API
+
+## Endpoint
+
+```
+GET /api/analytics/v2/brand/:name/intelligence
+```
+
+## Query Parameters
+
+| Param | Type | Default | Description |
+|-------|------|---------|-------------|
+| `window` | `7d\|30d\|90d` | `30d` | Time window for trend calculations |
+| `state` | string | - | Filter by state code (e.g., `AZ`) |
+| `category` | string | - | Filter by category (e.g., `Flower`) |
+
+## Response Payload Schema
+
+```typescript
+interface BrandIntelligenceResult {
+  brand_name: string;
+  window: '7d' | '30d' | '90d';
+  generated_at: string;  // ISO timestamp when data was computed
+
+  performance_snapshot: PerformanceSnapshot;
+  alerts: Alerts;
+  sku_performance: SkuPerformance[];
+  retail_footprint: RetailFootprint;
+  competitive_landscape: CompetitiveLandscape;
+  inventory_health: InventoryHealth;
+  promo_performance: PromoPerformance;
+}
+```
+
+---
+
+## Section 1: Performance Snapshot
+
+Summary cards with key brand metrics.
+
+```typescript
+interface PerformanceSnapshot {
+  active_skus: number;              // Total products in catalog
+  total_revenue_30d: number | null; // Estimated from qty × price
+  total_stores: number;             // Active retail partners
+  new_stores_30d: number;           // New distribution in window
+  market_share: number | null;      // % of category SKUs
+  avg_wholesale_price: number | null;
+  price_position: 'premium' | 'value' | 'competitive';
+}
+```
+
+**UI Label Mapping:**
+| Field | User-Facing Label | Helper Text |
+|-------|-------------------|-------------|
+| `active_skus` | Active Products | X total in catalog |
+| `total_revenue_30d` | Monthly Revenue | Estimated from sales |
+| `total_stores` | Retail Distribution | Active retail partners |
+| `new_stores_30d` | New Opportunities | X new in last 30 days |
+| `market_share` | Category Position | % of category |
+| `avg_wholesale_price` | Avg Wholesale | Per unit |
+| `price_position` | Pricing Tier | Premium/Value/Market Rate |
+
+---
+
+## Section 2: Alerts
+
+Issues requiring attention.
+
+```typescript
+interface Alerts {
+  lost_stores_30d_count: number;
+  lost_skus_30d_count: number;
+  competitor_takeover_count: number;
+  avg_oos_duration_days: number | null;
+  avg_reorder_lag_days: number | null;
+  items: AlertItem[];
+}
+
+interface AlertItem {
+  type: 'lost_store' | 'delisted_sku' | 'shelf_loss' | 'extended_oos';
+  severity: 'critical' | 'warning';
+  store_name?: string;
+  product_name?: string;
+  competitor_brand?: string;
+  days_since?: number;
+  state_code?: string;
+}
+```
+
+**UI Label Mapping:**
+| Field | User-Facing Label |
+|-------|-------------------|
+| `lost_stores_30d_count` | Accounts at Risk |
+| `lost_skus_30d_count` | Delisted SKUs |
+| `competitor_takeover_count` | Shelf Losses |
+| `avg_oos_duration_days` | Avg Stockout Length |
+| `avg_reorder_lag_days` | Avg Restock Time |
+| `severity: critical` | Urgent |
+| `severity: warning` | Watch |
+
+---
+
+## Section 3: SKU Performance (Product Velocity)
+
+How fast each SKU sells.
+
+```typescript
+interface SkuPerformance {
+  store_product_id: number;
+  product_name: string;
+  category: string | null;
+  daily_velocity: number;        // Units/day estimate
+  velocity_status: 'hot' | 'steady' | 'slow' | 'stale';
+  retail_price: number | null;
+  on_sale: boolean;
+  stores_carrying: number;
+  stock_status: 'in_stock' | 'low_stock' | 'out_of_stock';
+}
+```
+
+**UI Label Mapping:**
+| Field | User-Facing Label |
+|-------|-------------------|
+| `daily_velocity` | Daily Rate |
+| `velocity_status` | Momentum |
+| `velocity_status: hot` | Hot |
+| `velocity_status: steady` | Steady |
+| `velocity_status: slow` | Slow |
+| `velocity_status: stale` | Stale |
+| `retail_price` | Retail Price |
+| `on_sale` | Promo (badge) |
+
+**Velocity Thresholds:**
+- `hot`: >= 5 units/day
+- `steady`: >= 1 unit/day
+- `slow`: >= 0.1 units/day
+- `stale`: < 0.1 units/day
+
+---
+
+## Section 4: Retail Footprint
+
+Store placement and coverage.
+
+```typescript
+interface RetailFootprint {
+  total_stores: number;
+  in_stock_count: number;
+  out_of_stock_count: number;
+  penetration_by_region: RegionPenetration[];
+  whitespace_stores: WhitespaceStore[];
+}
+
+interface RegionPenetration {
+  state_code: string;
+  store_count: number;
+  percent_reached: number;    // % of state's dispensaries
+  in_stock: number;
+  out_of_stock: number;
+}
+
+interface WhitespaceStore {
+  store_id: number;
+  store_name: string;
+  state_code: string;
+  city: string | null;
+  category_fit: number;       // How many competing brands they carry
+  competitor_brands: string[];
+}
+```
+
+**UI Label Mapping:**
+| Field | User-Facing Label |
+|-------|-------------------|
+| `penetration_by_region` | Market Coverage by Region |
+| `percent_reached` | X% reached |
+| `in_stock` | X stocked |
+| `out_of_stock` | X out |
+| `whitespace_stores` | Expansion Opportunities |
+| `category_fit` | X fit |
+
+---
+
+## Section 5: Competitive Landscape
+
+Market positioning vs competitors.
+
+```typescript
+interface CompetitiveLandscape {
+  brand_price_position: 'premium' | 'value' | 'competitive';
+  market_share_trend: MarketSharePoint[];
+  competitors: Competitor[];
+  head_to_head_skus: HeadToHead[];
+}
+
+interface MarketSharePoint {
+  date: string;
+  share_percent: number;
+}
+
+interface Competitor {
+  brand_name: string;
+  store_overlap_percent: number;
+  price_position: 'premium' | 'value' | 'competitive';
+  avg_price: number | null;
+  sku_count: number;
+}
+
+interface HeadToHead {
+  product_name: string;
+  brand_price: number;
+  competitor_brand: string;
+  competitor_price: number;
+  price_diff_percent: number;
+}
+```
+
+**UI Label Mapping:**
+| Field | User-Facing Label |
+|-------|-------------------|
+| `price_position: premium` | Premium Tier |
+| `price_position: value` | Value Leader |
+| `price_position: competitive` | Market Rate |
+| `market_share_trend` | Share of Shelf Trend |
+| `head_to_head_skus` | Price Comparison |
+| `store_overlap_percent` | X% store overlap |
+
+---
+
+## Section 6: Inventory Health
+
+Stock projections and risk levels.
+
+```typescript
+interface InventoryHealth {
+  critical_count: number;      // <7 days stock
+  warning_count: number;       // 7-14 days stock
+  healthy_count: number;       // 14-90 days stock
+  overstocked_count: number;   // >90 days stock
+  skus: InventorySku[];
+  overstock_alert: OverstockItem[];
+}
+
+interface InventorySku {
+  store_product_id: number;
+  product_name: string;
+  store_name: string;
+  days_of_stock: number | null;
+  risk_level: 'critical' | 'elevated' | 'moderate' | 'healthy';
+  current_quantity: number | null;
+  daily_sell_rate: number | null;
+}
+
+interface OverstockItem {
+  product_name: string;
+  store_name: string;
+  excess_units: number;
+  days_of_stock: number;
+}
+```
+
+**UI Label Mapping:**
+| Field | User-Facing Label |
+|-------|-------------------|
+| `risk_level: critical` | Reorder Now |
+| `risk_level: elevated` | Low Stock |
+| `risk_level: moderate` | Monitor |
+| `risk_level: healthy` | Healthy |
+| `critical_count` | Urgent (<7 days) |
+| `warning_count` | Low (7-14 days) |
+| `overstocked_count` | Excess (>90 days) |
+| `days_of_stock` | X days remaining |
+| `overstock_alert` | Overstock Alert |
+| `excess_units` | X excess units |
+
+---
+
+## Section 7: Promotion Effectiveness
+
+How promotions impact sales.
+
+```typescript
+interface PromoPerformance {
+  avg_baseline_velocity: number | null;
+  avg_promo_velocity: number | null;
+  avg_velocity_lift: number | null;     // % increase during promo
+  avg_efficiency_score: number | null;  // ROI proxy
+  promotions: Promotion[];
+}
+
+interface Promotion {
+  product_name: string;
+  store_name: string;
+  status: 'active' | 'scheduled' | 'ended';
+  start_date: string;
+  end_date: string | null;
+  regular_price: number;
+  promo_price: number;
+  discount_percent: number;
+  baseline_velocity: number | null;
+  promo_velocity: number | null;
+  velocity_lift: number | null;
+  efficiency_score: number | null;
+}
+```
+
+**UI Label Mapping:**
+| Field | User-Facing Label |
+|-------|-------------------|
+| `avg_baseline_velocity` | Normal Rate |
+| `avg_promo_velocity` | During Promos |
+| `avg_velocity_lift` | Avg Sales Lift |
+| `avg_efficiency_score` | ROI Score |
+| `velocity_lift` | Sales Lift |
+| `efficiency_score` | ROI Score |
+| `status: active` | Live |
+| `status: scheduled` | Scheduled |
+| `status: ended` | Ended |
+
+---
+
+## Example Queries
+
+### Get full payload
+```javascript
+const response = await fetch('/api/analytics/v2/brand/Wyld/intelligence?window=30d');
+const data = await response.json();
+```
+
+### Extract summary cards (flattened)
+```javascript
+const { performance_snapshot: ps, alerts } = data;
+
+const summaryCards = {
+  activeProducts: ps.active_skus,
+  monthlyRevenue: ps.total_revenue_30d,
+  retailDistribution: ps.total_stores,
+  newOpportunities: ps.new_stores_30d,
+  categoryPosition: ps.market_share,
+  avgWholesale: ps.avg_wholesale_price,
+  pricingTier: ps.price_position,
+  accountsAtRisk: alerts.lost_stores_30d_count,
+  delistedSkus: alerts.lost_skus_30d_count,
+  shelfLosses: alerts.competitor_takeover_count,
+};
+```
+
+### Get top 10 fastest selling SKUs
+```javascript
+const topSkus = data.sku_performance
+  .filter(sku => sku.velocity_status === 'hot' || sku.velocity_status === 'steady')
+  .sort((a, b) => b.daily_velocity - a.daily_velocity)
+  .slice(0, 10);
+```
+
+### Get critical inventory alerts only
+```javascript
+const criticalInventory = data.inventory_health.skus
+  .filter(sku => sku.risk_level === 'critical');
+```
+
+### Get states with <50% penetration
+```javascript
+const underPenetrated = data.retail_footprint.penetration_by_region
+  .filter(region => region.percent_reached < 50)
+  .sort((a, b) => a.percent_reached - b.percent_reached);
+```
+
+### Get active promotions with positive lift
+```javascript
+const effectivePromos = data.promo_performance.promotions
+  .filter(p => p.status === 'active' && p.velocity_lift > 0)
+  .sort((a, b) => b.velocity_lift - a.velocity_lift);
+```
+
+### Build chart data for market share trend
+```javascript
+const chartData = data.competitive_landscape.market_share_trend.map(point => ({
+  x: new Date(point.date),
+  y: point.share_percent,
+}));
+```
+
+---
+
+## Notes for Frontend Implementation
+
+1. **All fields are snake_case** - transform to camelCase if needed
+2. **Null values are possible** - handle gracefully in UI
+3. **Arrays may be empty** - show appropriate empty states
+4. **Timestamps are ISO format** - parse with `new Date()`
+5. **Percentages are already computed** - no need to multiply by 100
+6. **The `window` parameter affects trend calculations** - 7d/30d/90d

backend/docs/_archive/CRAWL_PIPELINE.md

@@ -0,0 +1,539 @@
+# Crawl Pipeline Documentation
+
+## Overview
+
+The crawl pipeline fetches product data from Dutchie dispensary menus and stores it in the canonical database. This document covers the complete flow from task scheduling to data storage.
+
+---
+
+## Pipeline Stages
+
+```
+┌─────────────────────┐
+│  store_discovery    │  Find new dispensaries
+└─────────┬───────────┘
+          │
+          ▼
+┌─────────────────────┐
+│ entry_point_discovery│  Resolve slug → platform_dispensary_id
+└─────────┬───────────┘
+          │
+          ▼
+┌─────────────────────┐
+│  product_discovery  │  Initial product crawl
+└─────────┬───────────┘
+          │
+          ▼
+┌─────────────────────┐
+│   product_resync    │  Recurring crawl (every 4 hours)
+└─────────────────────┘
+```
+
+---
+
+## Stage Details
+
+### 1. Store Discovery
+**Purpose:** Find new dispensaries to crawl
+
+**Handler:** `src/tasks/handlers/store-discovery.ts`
+
+**Flow:**
+1. Query Dutchie `ConsumerDispensaries` GraphQL for cities/states
+2. Extract dispensary info (name, address, menu_url)
+3. Insert into `dutchie_discovery_locations`
+4. Queue `entry_point_discovery` for each new location
+
+---
+
+### 2. Entry Point Discovery
+**Purpose:** Resolve menu URL slug to platform_dispensary_id (MongoDB ObjectId)
+
+**Handler:** `src/tasks/handlers/entry-point-discovery.ts`
+
+**Flow:**
+1. Load dispensary from database
+2. Extract slug from `menu_url`:
+   - `/embedded-menu/<slug>` or `/dispensary/<slug>`
+3. Start stealth session (fingerprint + proxy)
+4. Query `resolveDispensaryIdWithDetails(slug)` via GraphQL
+5. Update dispensary with `platform_dispensary_id`
+6. Queue `product_discovery` task
+
+**Example:**
+```
+menu_url: https://dutchie.com/embedded-menu/deeply-rooted
+slug: deeply-rooted
+platform_dispensary_id: 6405ef617056e8014d79101b
+```
+
+---
+
+### 3. Product Discovery
+**Purpose:** Initial crawl of a new dispensary
+
+**Handler:** `src/tasks/handlers/product-discovery.ts`
+
+Same as product_resync but for first-time crawls.
+
+---
+
+### 4. Product Resync
+**Purpose:** Recurring crawl to capture price/stock changes
+
+**Handler:** `src/tasks/handlers/product-resync.ts`
+
+**Flow:**
+
+#### Step 1: Load Dispensary Info
+```sql
+SELECT id, name, platform_dispensary_id, menu_url, state
+FROM dispensaries
+WHERE id = $1 AND crawl_enabled = true
+```
+
+#### Step 2: Start Stealth Session
+- Generate random browser fingerprint
+- Set locale/timezone matching state
+- Optional proxy rotation
+
+#### Step 3: Fetch Products via GraphQL
+**Endpoint:** `https://dutchie.com/api-3/graphql`
+
+**Variables:**
+```javascript
+{
+  includeEnterpriseSpecials: false,
+  productsFilter: {
+    dispensaryId: "<platform_dispensary_id>",
+    pricingType: "rec",
+    Status: "All",
+    types: [],
+    useCache: false,
+    isDefaultSort: true,
+    sortBy: "popularSortIdx",
+    sortDirection: 1,
+    bypassOnlineThresholds: true,
+    isKioskMenu: false,
+    removeProductsBelowOptionThresholds: false
+  },
+  page: 0,
+  perPage: 100
+}
+```
+
+**Key Notes:**
+- `Status: "All"` returns all products (Active returns same count)
+- `Status: null` returns 0 products (broken)
+- `pricingType: "rec"` returns BOTH rec and med prices
+- Paginate until `products.length < perPage` or `allProducts.length >= totalCount`
+
+#### Step 4: Normalize Data
+Transform raw Dutchie payload to canonical format via `DutchieNormalizer`.
+
+#### Step 5: Upsert Products
+Insert/update `store_products` table with normalized data.
+
+#### Step 6: Create Snapshots
+Insert point-in-time record to `store_product_snapshots`.
+
+#### Step 7: Track Missing Products (OOS Detection)
+```sql
+-- Reset consecutive_misses for products IN the feed
+UPDATE store_products
+SET consecutive_misses = 0, last_seen_at = NOW()
+WHERE dispensary_id = $1
+  AND provider = 'dutchie'
+  AND provider_product_id = ANY($2)
+
+-- Increment for products NOT in feed
+UPDATE store_products
+SET consecutive_misses = consecutive_misses + 1
+WHERE dispensary_id = $1
+  AND provider = 'dutchie'
+  AND provider_product_id NOT IN (...)
+  AND consecutive_misses < 3
+
+-- Mark OOS at 3 consecutive misses
+UPDATE store_products
+SET stock_status = 'oos', is_in_stock = false
+WHERE dispensary_id = $1
+  AND consecutive_misses >= 3
+  AND stock_status != 'oos'
+```
+
+#### Step 8: Download Images
+For new products, download and store images locally.
+
+#### Step 9: Update Dispensary
+```sql
+UPDATE dispensaries SET last_crawl_at = NOW() WHERE id = $1
+```
+
+---
+
+## GraphQL Payload Structure
+
+### Product Fields (from filteredProducts.products[])
+
+| Field | Type | Description |
+|-------|------|-------------|
+| `_id` / `id` | string | MongoDB ObjectId (24 hex chars) |
+| `Name` | string | Product display name |
+| `brandName` | string | Brand name |
+| `brand.name` | string | Brand name (nested) |
+| `brand.description` | string | Brand description |
+| `type` | string | Category (Flower, Edible, Concentrate, etc.) |
+| `subcategory` | string | Subcategory |
+| `strainType` | string | Hybrid, Indica, Sativa, N/A |
+| `Status` | string | Always "Active" in feed |
+| `Image` | string | Primary image URL |
+| `images[]` | array | All product images |
+
+### Pricing Fields
+
+| Field | Type | Description |
+|-------|------|-------------|
+| `Prices[]` | number[] | Rec prices per option |
+| `recPrices[]` | number[] | Rec prices |
+| `medicalPrices[]` | number[] | Medical prices |
+| `recSpecialPrices[]` | number[] | Rec sale prices |
+| `medicalSpecialPrices[]` | number[] | Medical sale prices |
+| `Options[]` | string[] | Size options ("1/8oz", "1g", etc.) |
+| `rawOptions[]` | string[] | Raw weight options ("3.5g") |
+
+### Inventory Fields (POSMetaData.children[])
+
+| Field | Type | Description |
+|-------|------|-------------|
+| `quantity` | number | Total inventory count |
+| `quantityAvailable` | number | Available for online orders |
+| `kioskQuantityAvailable` | number | Available for kiosk orders |
+| `option` | string | Which size option this is for |
+
+### Potency Fields
+
+| Field | Type | Description |
+|-------|------|-------------|
+| `THCContent.range[]` | number[] | THC percentage |
+| `CBDContent.range[]` | number[] | CBD percentage |
+| `cannabinoidsV2[]` | array | Detailed cannabinoid breakdown |
+
+### Specials (specialData.bogoSpecials[])
+
+| Field | Type | Description |
+|-------|------|-------------|
+| `specialName` | string | Deal name |
+| `specialType` | string | "bogo", "sale", etc. |
+| `itemsForAPrice.value` | string | Bundle price |
+| `bogoRewards[].totalQuantity.quantity` | number | Required quantity |
+
+---
+
+## OOS Detection Logic
+
+Products disappear from the Dutchie feed when they go out of stock. We track this via `consecutive_misses`:
+
+| Scenario | Action |
+|----------|--------|
+| Product in feed | `consecutive_misses = 0` |
+| Product missing 1st time | `consecutive_misses = 1` |
+| Product missing 2nd time | `consecutive_misses = 2` |
+| Product missing 3rd time | `consecutive_misses = 3`, mark `stock_status = 'oos'` |
+| Product returns to feed | `consecutive_misses = 0`, update stock_status |
+
+**Why 3 misses?**
+- Protects against false positives from crawl failures
+- Single bad crawl doesn't trigger mass OOS alerts
+- Balances detection speed vs accuracy
+
+---
+
+## Database Tables
+
+### store_products
+Current state of each product:
+- `provider_product_id` - Dutchie's MongoDB ObjectId
+- `name_raw`, `brand_name_raw` - Raw values from feed
+- `price_rec`, `price_med` - Current prices
+- `is_in_stock`, `stock_status` - Availability
+- `consecutive_misses` - OOS detection counter
+- `last_seen_at` - Last time product was in feed
+
+### store_product_snapshots
+Point-in-time records for historical analysis:
+- One row per product per crawl
+- Captures price, stock, potency at that moment
+- Used for price history, analytics
+
+### dispensaries
+Store metadata:
+- `platform_dispensary_id` - MongoDB ObjectId for GraphQL
+- `menu_url` - Source URL
+- `last_crawl_at` - Last successful crawl
+- `crawl_enabled` - Whether to crawl
+
+---
+
+## Worker Roles
+
+Workers pull tasks from the `worker_tasks` queue based on their assigned role.
+
+| Role | Name | Description | Handler |
+|------|------|-------------|---------|
+| `product_resync` | Product Resync | Re-crawl dispensary products for price/stock changes | `handleProductResync` |
+| `product_discovery` | Product Discovery | Initial product discovery for new dispensaries | `handleProductDiscovery` |
+| `store_discovery` | Store Discovery | Discover new dispensary locations | `handleStoreDiscovery` |
+| `entry_point_discovery` | Entry Point Discovery | Resolve platform IDs from menu URLs | `handleEntryPointDiscovery` |
+| `analytics_refresh` | Analytics Refresh | Refresh materialized views and analytics | `handleAnalyticsRefresh` |
+
+**API Endpoint:** `GET /api/worker-registry/roles`
+
+---
+
+## Scheduling
+
+Crawls are scheduled via `worker_tasks` table:
+
+| Role | Frequency | Description |
+|------|-----------|-------------|
+| `product_resync` | Every 4 hours | Regular product refresh |
+| `product_discovery` | On-demand | First crawl for new stores |
+| `entry_point_discovery` | On-demand | New store setup |
+| `store_discovery` | Daily | Find new stores |
+| `analytics_refresh` | Daily | Refresh analytics materialized views |
+
+---
+
+## Priority & On-Demand Tasks
+
+Tasks are claimed by workers in order of **priority DESC, created_at ASC**.
+
+### Priority Levels
+
+| Priority | Use Case | Example |
+|----------|----------|---------|
+| 0 | Scheduled/batch tasks | Daily product_resync generation |
+| 10 | On-demand/chained tasks | entry_point → product_discovery |
+| Higher | Urgent/manual triggers | Admin-triggered immediate crawl |
+
+### Task Chaining
+
+When a task completes, the system automatically creates follow-up tasks:
+
+```
+store_discovery (completed)
+    └─► entry_point_discovery (priority: 10) for each new store
+
+entry_point_discovery (completed, success)
+    └─► product_discovery (priority: 10) for that store
+
+product_discovery (completed)
+    └─► [no chain] Store enters regular resync schedule
+```
+
+### On-Demand Task Creation
+
+Use the task service to create high-priority tasks:
+
+```typescript
+// Create immediate product resync for a store
+await taskService.createTask({
+  role: 'product_resync',
+  dispensary_id: 123,
+  platform: 'dutchie',
+  priority: 20, // Higher than batch tasks
+});
+
+// Convenience methods with default high priority (10)
+await taskService.createEntryPointTask(dispensaryId, 'dutchie');
+await taskService.createProductDiscoveryTask(dispensaryId, 'dutchie');
+await taskService.createStoreDiscoveryTask('dutchie', 'AZ');
+```
+
+### Claim Function
+
+The `claim_task()` SQL function atomically claims tasks:
+- Respects priority ordering (higher = first)
+- Uses `FOR UPDATE SKIP LOCKED` for concurrency
+- Prevents multiple active tasks per store
+
+---
+
+## Image Storage
+
+Images are downloaded from Dutchie's AWS S3 and stored locally with on-demand resizing.
+
+### Storage Path
+```
+/storage/images/products/<state>/<store>/<brand>/<product_id>/image-<hash>.webp
+/storage/images/brands/<brand>/logo-<hash>.webp
+```
+
+**Example:**
+```
+/storage/images/products/az/az-deeply-rooted/bud-bros/6913e3cd444eac3935e928b9/image-ae38b1f9.webp
+```
+
+### Image Proxy API
+Served via `/img/*` with on-demand resizing using **sharp**:
+
+```
+GET /img/products/az/az-deeply-rooted/bud-bros/6913e3cd444eac3935e928b9/image-ae38b1f9.webp?w=200
+```
+
+| Param | Description |
+|-------|-------------|
+| `w` | Width in pixels (max 4000) |
+| `h` | Height in pixels (max 4000) |
+| `q` | Quality 1-100 (default 80) |
+| `fit` | cover, contain, fill, inside, outside |
+| `blur` | Blur sigma (0.3-1000) |
+| `gray` | Grayscale (1 = enabled) |
+| `format` | webp, jpeg, png, avif (default webp) |
+
+### Key Files
+| File | Purpose |
+|------|---------|
+| `src/utils/image-storage.ts` | Download & save images to local filesystem |
+| `src/routes/image-proxy.ts` | On-demand resize/transform at `/img/*` |
+
+### Download Rules
+
+| Scenario | Image Action |
+|----------|--------------|
+| **New product (first crawl)** | Download if `primaryImageUrl` exists |
+| **Existing product (refresh)** | Download only if `local_image_path` is NULL (backfill) |
+| **Product already has local image** | Skip download entirely |
+
+**Logic:**
+- Images are downloaded **once** and never re-downloaded on subsequent crawls
+- `skipIfExists: true` - filesystem check prevents re-download even if queued
+- First crawl: all products get images
+- Refresh crawl: only new products or products missing local images
+
+### Storage Rules
+- **NO MinIO** - local filesystem only (`STORAGE_DRIVER=local`)
+- Store full resolution, resize on-demand via `/img` proxy
+- Convert to webp for consistency using **sharp**
+- Preserve original Dutchie URL as fallback in `image_url` column
+- Local path stored in `local_image_path` column
+
+---
+
+## Stealth & Anti-Detection
+
+**PROXIES ARE REQUIRED** - Workers will fail to start if no active proxies are available in the database. All HTTP requests to Dutchie go through a proxy.
+
+Workers automatically initialize anti-detection systems on startup.
+
+### Components
+
+| Component | Purpose | Source |
+|-----------|---------|--------|
+| **CrawlRotator** | Coordinates proxy + UA rotation | `src/services/crawl-rotator.ts` |
+| **ProxyRotator** | Round-robin proxy selection, health tracking | `src/services/crawl-rotator.ts` |
+| **UserAgentRotator** | Cycles through realistic browser fingerprints | `src/services/crawl-rotator.ts` |
+| **Dutchie Client** | Curl-based HTTP with auto-retry on 403 | `src/platforms/dutchie/client.ts` |
+
+### Initialization Flow
+
+```
+Worker Start
+    │
+    ├─► initializeStealth()
+    │       │
+    │       ├─► CrawlRotator.initialize()
+    │       │       └─► Load proxies from `proxies` table
+    │       │
+    │       └─► setCrawlRotator(rotator)
+    │               └─► Wire to Dutchie client
+    │
+    └─► Process tasks...
+```
+
+### Stealth Session (per task)
+
+Each crawl task starts a stealth session:
+
+```typescript
+// In product-refresh.ts, entry-point-discovery.ts
+const session = startSession(dispensary.state || 'AZ', 'America/Phoenix');
+```
+
+This creates a new identity with:
+- **Random fingerprint:** Chrome/Firefox/Safari/Edge on Win/Mac/Linux
+- **Accept-Language:** Matches timezone (e.g., `America/Phoenix` → `en-US,en;q=0.9`)
+- **sec-ch-ua headers:** Proper Client Hints for the browser profile
+
+### On 403 Block
+
+When Dutchie returns 403, the client automatically:
+
+1. Records failure on current proxy (increments `failure_count`)
+2. If proxy has 5+ failures, deactivates it
+3. Rotates to next healthy proxy
+4. Rotates fingerprint
+5. Retries the request
+
+### Proxy Table Schema
+
+```sql
+CREATE TABLE proxies (
+  id SERIAL PRIMARY KEY,
+  host VARCHAR(255) NOT NULL,
+  port INTEGER NOT NULL,
+  username VARCHAR(100),
+  password VARCHAR(100),
+  protocol VARCHAR(10) DEFAULT 'http',  -- http, https, socks5
+  is_active BOOLEAN DEFAULT true,
+  last_used_at TIMESTAMPTZ,
+  failure_count INTEGER DEFAULT 0,
+  success_count INTEGER DEFAULT 0,
+  avg_response_time_ms INTEGER,
+  last_failure_at TIMESTAMPTZ,
+  last_error TEXT
+);
+```
+
+### Configuration
+
+Proxies are mandatory. There is no environment variable to disable them. Workers will refuse to start without active proxies in the database.
+
+### User-Agent Generation
+
+See `workflow-12102025.md` for full specification.
+
+**Summary:**
+- Uses `intoli/user-agents` library (daily-updated market share data)
+- Device distribution: Mobile 62%, Desktop 36%, Tablet 2%
+- Browser whitelist: Chrome, Safari, Edge, Firefox only
+- UA sticks until IP rotates (403 or manual rotation)
+- Failure = alert admin + stop crawl (no fallback)
+
+Each fingerprint includes proper `sec-ch-ua`, `sec-ch-ua-platform`, and `sec-ch-ua-mobile` headers.
+
+---
+
+## Error Handling
+
+- **GraphQL errors:** Logged, task marked failed, retried later
+- **Normalization errors:** Logged as warnings, continue with valid products
+- **Image download errors:** Non-fatal, logged, continue
+- **Database errors:** Task fails, will be retried
+- **403 blocks:** Auto-rotate proxy + fingerprint, retry (up to 3 retries)
+
+---
+
+## Files
+
+| File | Purpose |
+|------|---------|
+| `src/tasks/handlers/product-resync.ts` | Main crawl handler |
+| `src/tasks/handlers/entry-point-discovery.ts` | Slug → ID resolution |
+| `src/platforms/dutchie/index.ts` | GraphQL client, session management |
+| `src/hydration/normalizers/dutchie.ts` | Payload normalization |
+| `src/hydration/canonical-upsert.ts` | Database upsert logic |
+| `src/utils/image-storage.ts` | Image download and local storage |
+| `src/routes/image-proxy.ts` | On-demand image resizing |
+| `migrations/075_consecutive_misses.sql` | OOS tracking column |

backend/docs/_archive/ORGANIC_SCRAPING_GUIDE.md

@@ -0,0 +1,297 @@
+# Organic Browser-Based Scraping Guide
+
+**Last Updated:** 2025-12-12
+**Status:** Production-ready proof of concept
+
+---
+
+## Overview
+
+This document describes the "organic" browser-based approach to scraping Dutchie dispensary menus. Unlike direct curl/axios requests, this method uses a real browser session to make API calls, making requests appear natural and reducing detection risk.
+
+---
+
+## Why Organic Scraping?
+
+| Approach | Detection Risk | Speed | Complexity |
+|----------|---------------|-------|------------|
+| Direct curl | Higher | Fast | Low |
+| curl-impersonate | Medium | Fast | Medium |
+| **Browser-based (organic)** | **Lowest** | Slower | Higher |
+
+Direct curl requests can be fingerprinted via:
+- TLS fingerprint (cipher suites, extensions)
+- Header order and values
+- Missing cookies/session data
+- Request patterns
+
+Browser-based requests inherit:
+- Real Chrome TLS fingerprint
+- Session cookies from page visit
+- Natural header order
+- JavaScript execution environment
+
+---
+
+## Implementation
+
+### Dependencies
+
+```bash
+npm install puppeteer puppeteer-extra puppeteer-extra-plugin-stealth
+```
+
+### Core Script: `test-intercept.js`
+
+Located at: `backend/test-intercept.js`
+
+```javascript
+const puppeteer = require('puppeteer-extra');
+const StealthPlugin = require('puppeteer-extra-plugin-stealth');
+const fs = require('fs');
+
+puppeteer.use(StealthPlugin());
+
+async function capturePayload(config) {
+  const { dispensaryId, platformId, cName, outputPath } = config;
+
+  const browser = await puppeteer.launch({
+    headless: 'new',
+    args: ['--no-sandbox', '--disable-setuid-sandbox']
+  });
+
+  const page = await browser.newPage();
+
+  // STEP 1: Establish session by visiting the menu
+  const embedUrl = `https://dutchie.com/embedded-menu/${cName}?menuType=rec`;
+  await page.goto(embedUrl, { waitUntil: 'networkidle2', timeout: 60000 });
+
+  // STEP 2: Fetch ALL products using GraphQL from browser context
+  const result = await page.evaluate(async (platformId) => {
+    const allProducts = [];
+    let pageNum = 0;
+    const perPage = 100;
+    let totalCount = 0;
+    const sessionId = 'browser-session-' + Date.now();
+
+    while (pageNum < 30) {
+      const variables = {
+        includeEnterpriseSpecials: false,
+        productsFilter: {
+          dispensaryId: platformId,
+          pricingType: 'rec',
+          Status: 'Active',  // CRITICAL: Must be 'Active', not null
+          types: [],
+          useCache: true,
+          isDefaultSort: true,
+          sortBy: 'popularSortIdx',
+          sortDirection: 1,
+          bypassOnlineThresholds: true,
+          isKioskMenu: false,
+          removeProductsBelowOptionThresholds: false,
+        },
+        page: pageNum,
+        perPage: perPage,
+      };
+
+      const extensions = {
+        persistedQuery: {
+          version: 1,
+          sha256Hash: 'ee29c060826dc41c527e470e9ae502c9b2c169720faa0a9f5d25e1b9a530a4a0'
+        }
+      };
+
+      const qs = new URLSearchParams({
+        operationName: 'FilteredProducts',
+        variables: JSON.stringify(variables),
+        extensions: JSON.stringify(extensions)
+      });
+
+      const response = await fetch(`https://dutchie.com/api-3/graphql?${qs}`, {
+        method: 'GET',
+        headers: {
+          'Accept': 'application/json',
+          'content-type': 'application/json',
+          'x-dutchie-session': sessionId,
+          'apollographql-client-name': 'Marketplace (production)',
+        },
+        credentials: 'include'
+      });
+
+      const json = await response.json();
+      const data = json?.data?.filteredProducts;
+      if (!data?.products) break;
+
+      allProducts.push(...data.products);
+      if (pageNum === 0) totalCount = data.queryInfo?.totalCount || 0;
+      if (allProducts.length >= totalCount) break;
+
+      pageNum++;
+      await new Promise(r => setTimeout(r, 200)); // Polite delay
+    }
+
+    return { products: allProducts, totalCount };
+  }, platformId);
+
+  await browser.close();
+
+  // STEP 3: Save payload
+  const payload = {
+    dispensaryId,
+    platformId,
+    cName,
+    fetchedAt: new Date().toISOString(),
+    productCount: result.products.length,
+    products: result.products,
+  };
+
+  fs.writeFileSync(outputPath, JSON.stringify(payload, null, 2));
+  return payload;
+}
+```
+
+---
+
+## Critical Parameters
+
+### GraphQL Hash (FilteredProducts)
+
+```
+ee29c060826dc41c527e470e9ae502c9b2c169720faa0a9f5d25e1b9a530a4a0
+```
+
+**WARNING:** Using the wrong hash returns HTTP 400.
+
+### Status Parameter
+
+| Value | Result |
+|-------|--------|
+| `'Active'` | Returns in-stock products (1019 in test) |
+| `null` | Returns 0 products |
+| `'All'` | Returns HTTP 400 |
+
+**ALWAYS use `Status: 'Active'`**
+
+### Required Headers
+
+```javascript
+{
+  'Accept': 'application/json',
+  'content-type': 'application/json',
+  'x-dutchie-session': 'unique-session-id',
+  'apollographql-client-name': 'Marketplace (production)',
+}
+```
+
+### Endpoint
+
+```
+https://dutchie.com/api-3/graphql
+```
+
+---
+
+## Performance Benchmarks
+
+Test store: AZ-Deeply-Rooted (1019 products)
+
+| Metric | Value |
+|--------|-------|
+| Total products | 1019 |
+| Time | 18.5 seconds |
+| Payload size | 11.8 MB |
+| Pages fetched | 11 (100 per page) |
+| Success rate | 100% |
+
+---
+
+## Payload Format
+
+The output matches the existing `payload-fetch.ts` handler format:
+
+```json
+{
+  "dispensaryId": 123,
+  "platformId": "6405ef617056e8014d79101b",
+  "cName": "AZ-Deeply-Rooted",
+  "fetchedAt": "2025-12-12T05:05:19.837Z",
+  "productCount": 1019,
+  "products": [
+    {
+      "id": "6927508db4851262f629a869",
+      "Name": "Product Name",
+      "brand": { "name": "Brand Name", ... },
+      "type": "Flower",
+      "THC": "25%",
+      "Prices": [...],
+      "Options": [...],
+      ...
+    }
+  ]
+}
+```
+
+---
+
+## Integration Points
+
+### As a Task Handler
+
+The organic approach can be integrated as an alternative to curl-based fetching:
+
+```typescript
+// In src/tasks/handlers/organic-payload-fetch.ts
+export async function handleOrganicPayloadFetch(ctx: TaskContext): Promise<TaskResult> {
+  // Use puppeteer-based capture
+  // Save to same payload storage
+  // Queue product_refresh task
+}
+```
+
+### Worker Configuration
+
+Add to job_schedules:
+```sql
+INSERT INTO job_schedules (name, role, cron_expression)
+VALUES ('organic_product_crawl', 'organic_payload_fetch', '0 */6 * * *');
+```
+
+---
+
+## Troubleshooting
+
+### HTTP 400 Bad Request
+- Check hash is correct: `ee29c060...`
+- Verify Status is `'Active'` (string, not null)
+
+### 0 Products Returned
+- Status was likely `null` or `'All'` - use `'Active'`
+- Check platformId is valid MongoDB ObjectId
+
+### Session Not Established
+- Increase timeout on initial page.goto()
+- Check cName is valid (matches embedded-menu URL)
+
+### Detection/Blocking
+- StealthPlugin should handle most cases
+- Add random delays between pages
+- Use headless: 'new' (not true/false)
+
+---
+
+## Files Reference
+
+| File | Purpose |
+|------|---------|
+| `backend/test-intercept.js` | Proof of concept script |
+| `backend/src/platforms/dutchie/client.ts` | GraphQL hashes, curl implementation |
+| `backend/src/tasks/handlers/payload-fetch.ts` | Current curl-based handler |
+| `backend/src/utils/payload-storage.ts` | Payload save/load utilities |
+
+---
+
+## See Also
+
+- `DUTCHIE_CRAWL_WORKFLOW.md` - Full crawl pipeline documentation
+- `TASK_WORKFLOW_2024-12-10.md` - Task system architecture
+- `CLAUDE.md` - Project rules and constraints

backend/docs/_archive/README.md

@@ -0,0 +1,25 @@
+# ARCHIVED DOCUMENTATION
+
+**WARNING: These docs may be outdated or inaccurate.**
+
+The code has evolved significantly. These docs are kept for historical reference only.
+
+## What to Use Instead
+
+**The single source of truth is:**
+- `CLAUDE.md` (root) - Essential rules and quick reference
+- `docs/CODEBASE_MAP.md` - Current file/directory reference
+
+## Why Archive?
+
+These docs were written during development iterations and may reference:
+- Old file paths that no longer exist
+- Deprecated approaches (hydration, scraper-v2)
+- APIs that have changed
+- Database schemas that evolved
+
+## If You Need Details
+
+1. First check CODEBASE_MAP.md for current file locations
+2. Then read the actual source code
+3. Only use archive docs as a last resort for historical context

backend/docs/_archive/TASK_WORKFLOW_2024-12-10.md

@@ -0,0 +1,584 @@
+# Task Workflow Documentation
+**Date: 2024-12-10**
+
+This document describes the complete task/job processing architecture after the 2024-12-10 rewrite.
+
+---
+
+## Complete Architecture
+
+```
+┌─────────────────────────────────────────────────────────────────────────────────┐
+│                              KUBERNETES CLUSTER                                  │
+├─────────────────────────────────────────────────────────────────────────────────┤
+│                                                                                  │
+│  ┌─────────────────────────────────────────────────────────────────────────┐    │
+│  │                         API SERVER POD (scraper)                         │    │
+│  │                                                                          │    │
+│  │   ┌──────────────────┐     ┌────────────────────────────────────────┐   │    │
+│  │   │   Express API    │     │         TaskScheduler                   │   │    │
+│  │   │                  │     │   (src/services/task-scheduler.ts)      │   │    │
+│  │   │  /api/job-queue  │     │                                         │   │    │
+│  │   │  /api/tasks      │     │   • Polls every 60s                     │   │    │
+│  │   │  /api/schedules  │     │   • Checks task_schedules table         │   │    │
+│  │   └────────┬─────────┘     │   • SELECT FOR UPDATE SKIP LOCKED       │   │    │
+│  │            │               │   • Generates tasks when due            │   │    │
+│  │            │               └──────────────────┬─────────────────────┘   │    │
+│  │            │                                  │                          │    │
+│  └────────────┼──────────────────────────────────┼──────────────────────────┘    │
+│               │                                  │                               │
+│               │         ┌────────────────────────┘                               │
+│               │         │                                                        │
+│               ▼         ▼                                                        │
+│  ┌─────────────────────────────────────────────────────────────────────────┐    │
+│  │                          POSTGRESQL DATABASE                             │    │
+│  │                                                                          │    │
+│  │   ┌─────────────────────┐        ┌─────────────────────┐                │    │
+│  │   │   task_schedules    │        │    worker_tasks     │                │    │
+│  │   │                     │        │                     │                │    │
+│  │   │ • product_refresh   │───────►│ • pending tasks     │                │    │
+│  │   │ • store_discovery   │ create │ • claimed tasks     │                │    │
+│  │   │ • analytics_refresh │ tasks  │ • running tasks     │                │    │
+│  │   │                     │        │ • completed tasks   │                │    │
+│  │   │ next_run_at         │        │                     │                │    │
+│  │   │ last_run_at         │        │ role, dispensary_id │                │    │
+│  │   │ interval_hours      │        │ priority, status    │                │    │
+│  │   └─────────────────────┘        └──────────┬──────────┘                │    │
+│  │                                             │                            │    │
+│  └─────────────────────────────────────────────┼────────────────────────────┘    │
+│                                                │                                  │
+│                         ┌──────────────────────┘                                  │
+│                         │ Workers poll for tasks                                  │
+│                         │ (SELECT FOR UPDATE SKIP LOCKED)                         │
+│                         ▼                                                         │
+│  ┌─────────────────────────────────────────────────────────────────────────┐    │
+│  │                    WORKER PODS (StatefulSet: scraper-worker)             │    │
+│  │                                                                          │    │
+│  │   ┌─────────────┐  ┌─────────────┐  ┌─────────────┐  ┌─────────────┐    │    │
+│  │   │  Worker 0   │  │  Worker 1   │  │  Worker 2   │  │  Worker N   │    │    │
+│  │   │             │  │             │  │             │  │             │    │    │
+│  │   │ task-worker │  │ task-worker │  │ task-worker │  │ task-worker │    │    │
+│  │   │     .ts     │  │     .ts     │  │     .ts     │  │     .ts     │    │    │
+│  │   └─────────────┘  └─────────────┘  └─────────────┘  └─────────────┘    │    │
+│  │                                                                          │    │
+│  └──────────────────────────────────────────────────────────────────────────┘    │
+│                                                                                  │
+└──────────────────────────────────────────────────────────────────────────────────┘
+```
+
+---
+
+## Startup Sequence
+
+```
+┌─────────────────────────────────────────────────────────────────────────────┐
+│                        API SERVER STARTUP                                    │
+├─────────────────────────────────────────────────────────────────────────────┤
+│                                                                              │
+│   1. Express app initializes                                                 │
+│                    │                                                         │
+│                    ▼                                                         │
+│   2. runAutoMigrations()                                                     │
+│      • Runs pending migrations (including 079_task_schedules.sql)           │
+│                    │                                                         │
+│                    ▼                                                         │
+│   3. initializeMinio() / initializeImageStorage()                           │
+│                    │                                                         │
+│                    ▼                                                         │
+│   4. cleanupOrphanedJobs()                                                   │
+│                    │                                                         │
+│                    ▼                                                         │
+│   5. taskScheduler.start()  ◄─── NEW (per TASK_WORKFLOW_2024-12-10.md)      │
+│      │                                                                       │
+│      ├── Recover stale tasks (workers that died)                            │
+│      ├── Ensure default schedules exist in task_schedules                   │
+│      ├── Check and run any due schedules immediately                        │
+│      └── Start 60-second poll interval                                      │
+│                    │                                                         │
+│                    ▼                                                         │
+│   6. app.listen(PORT)                                                        │
+│                                                                              │
+└─────────────────────────────────────────────────────────────────────────────┘
+
+┌─────────────────────────────────────────────────────────────────────────────┐
+│                        WORKER POD STARTUP                                    │
+├─────────────────────────────────────────────────────────────────────────────┤
+│                                                                              │
+│   1. K8s starts pod from StatefulSet                                        │
+│                    │                                                         │
+│                    ▼                                                         │
+│   2. TaskWorker.constructor()                                               │
+│      • Create DB pool                                                        │
+│      • Create CrawlRotator                                                   │
+│                    │                                                         │
+│                    ▼                                                         │
+│   3. initializeStealth()                                                    │
+│      • Load proxies from DB (REQUIRED - fails if none)                      │
+│      • Wire rotator to Dutchie client                                       │
+│                    │                                                         │
+│                    ▼                                                         │
+│   4. register() with API                                                    │
+│      • Optional - continues if fails                                         │
+│                    │                                                         │
+│                    ▼                                                         │
+│   5. startRegistryHeartbeat() every 30s                                     │
+│                    │                                                         │
+│                    ▼                                                         │
+│   6. processNextTask() loop                                                 │
+│      │                                                                       │
+│      ├── Poll for pending task (FOR UPDATE SKIP LOCKED)                     │
+│      ├── Claim task atomically                                              │
+│      ├── Execute handler (product_refresh, store_discovery, etc.)           │
+│      ├── Mark complete/failed                                               │
+│      ├── Chain next task if applicable                                      │
+│      └── Loop                                                               │
+│                                                                              │
+└─────────────────────────────────────────────────────────────────────────────┘
+```
+
+---
+
+## Schedule Flow
+
+```
+┌─────────────────────────────────────────────────────────────────────────────┐
+│                     SCHEDULER POLL (every 60 seconds)                        │
+├─────────────────────────────────────────────────────────────────────────────┤
+│                                                                              │
+│   BEGIN TRANSACTION                                                          │
+│         │                                                                    │
+│         ▼                                                                    │
+│   SELECT * FROM task_schedules                                              │
+│   WHERE enabled = true AND next_run_at <= NOW()                             │
+│   FOR UPDATE SKIP LOCKED  ◄─── Prevents duplicate execution across replicas │
+│         │                                                                    │
+│         ▼                                                                    │
+│   For each due schedule:                                                     │
+│         │                                                                    │
+│         ├── product_refresh_all                                             │
+│         │   └─► Query dispensaries needing crawl                            │
+│         │   └─► Create product_refresh tasks in worker_tasks                │
+│         │                                                                    │
+│         ├── store_discovery_dutchie                                         │
+│         │   └─► Create single store_discovery task                          │
+│         │                                                                    │
+│         └── analytics_refresh                                                │
+│             └─► Create single analytics_refresh task                        │
+│         │                                                                    │
+│         ▼                                                                    │
+│   UPDATE task_schedules SET                                                  │
+│     last_run_at = NOW(),                                                     │
+│     next_run_at = NOW() + interval_hours                                    │
+│         │                                                                    │
+│         ▼                                                                    │
+│   COMMIT                                                                     │
+│                                                                              │
+└─────────────────────────────────────────────────────────────────────────────┘
+```
+
+---
+
+## Task Lifecycle
+
+```
+                                    ┌──────────┐
+                                    │ SCHEDULE │
+                                    │   DUE    │
+                                    └────┬─────┘
+                                         │
+                                         ▼
+┌──────────────┐    claim    ┌──────────────┐    start    ┌──────────────┐
+│   PENDING    │────────────►│   CLAIMED    │────────────►│   RUNNING    │
+└──────────────┘             └──────────────┘             └──────┬───────┘
+       ▲                                                        │
+       │                                         ┌──────────────┼──────────────┐
+       │ retry                                   │              │              │
+       │ (if retries < max)                      ▼              ▼              ▼
+       │                                  ┌──────────┐   ┌──────────┐   ┌──────────┐
+       └──────────────────────────────────│  FAILED  │   │ COMPLETED│   │  STALE   │
+                                          └──────────┘   └──────────┘   └────┬─────┘
+                                                                              │
+                                                              recover_stale_tasks()
+                                                                              │
+                                                                              ▼
+                                                                        ┌──────────┐
+                                                                        │ PENDING  │
+                                                                        └──────────┘
+```
+
+---
+
+## Database Tables
+
+### task_schedules (NEW - migration 079)
+
+Stores schedule definitions. Survives restarts.
+
+```sql
+CREATE TABLE task_schedules (
+  id SERIAL PRIMARY KEY,
+  name VARCHAR(100) NOT NULL UNIQUE,
+  role VARCHAR(50) NOT NULL,        -- product_refresh, store_discovery, etc.
+  enabled BOOLEAN DEFAULT TRUE,
+  interval_hours INTEGER NOT NULL,  -- How often to run
+  priority INTEGER DEFAULT 0,       -- Task priority when created
+  state_code VARCHAR(2),            -- Optional filter
+  last_run_at TIMESTAMPTZ,          -- When it last ran
+  next_run_at TIMESTAMPTZ,          -- When it's due next
+  last_task_count INTEGER,          -- Tasks created last run
+  last_error TEXT                   -- Error message if failed
+);
+```
+
+### worker_tasks (migration 074)
+
+The task queue. Workers pull from here.
+
+```sql
+CREATE TABLE worker_tasks (
+  id SERIAL PRIMARY KEY,
+  role task_role NOT NULL,          -- What type of work
+  dispensary_id INTEGER,            -- Which store (if applicable)
+  platform VARCHAR(50),             -- Which platform
+  status task_status DEFAULT 'pending',
+  priority INTEGER DEFAULT 0,       -- Higher = process first
+  scheduled_for TIMESTAMP,          -- Don't process before this time
+  worker_id VARCHAR(100),           -- Which worker claimed it
+  claimed_at TIMESTAMP,
+  started_at TIMESTAMP,
+  completed_at TIMESTAMP,
+  last_heartbeat_at TIMESTAMP,      -- For stale detection
+  result JSONB,
+  error_message TEXT,
+  retry_count INTEGER DEFAULT 0,
+  max_retries INTEGER DEFAULT 3
+);
+```
+
+---
+
+## Default Schedules
+
+| Name | Role | Interval | Priority | Description |
+|------|------|----------|----------|-------------|
+| `payload_fetch_all` | payload_fetch | 4 hours | 0 | Fetch payloads from Dutchie API (chains to product_refresh) |
+| `store_discovery_dutchie` | store_discovery | 24 hours | 5 | Find new Dutchie stores |
+| `analytics_refresh` | analytics_refresh | 6 hours | 0 | Refresh MVs |
+
+---
+
+## Task Roles
+
+| Role | Description | Creates Tasks For |
+|------|-------------|-------------------|
+| `payload_fetch` | **NEW** - Fetch from Dutchie API, save to disk | Each dispensary needing crawl |
+| `product_refresh` | **CHANGED** - Read local payload, normalize, upsert to DB | Chained from payload_fetch |
+| `store_discovery` | Find new dispensaries, returns newStoreIds[] | Single task per platform |
+| `entry_point_discovery` | **DEPRECATED** - Resolve platform IDs | No longer used |
+| `product_discovery` | Initial product fetch for new stores | Chained from store_discovery |
+| `analytics_refresh` | Refresh MVs | Single global task |
+
+### Payload/Refresh Separation (2024-12-10)
+
+The crawl workflow is now split into two phases:
+
+```
+payload_fetch (scheduled every 4h)
+  └─► Hit Dutchie GraphQL API
+  └─► Save raw JSON to /storage/payloads/{year}/{month}/{day}/store_{id}_{ts}.json.gz
+  └─► Record metadata in raw_crawl_payloads table
+  └─► Queue product_refresh task with payload_id
+
+product_refresh (chained from payload_fetch)
+  └─► Load payload from filesystem (NOT from API)
+  └─► Normalize via DutchieNormalizer
+  └─► Upsert to store_products
+  └─► Create snapshots
+  └─► Track missing products
+  └─► Download images
+```
+
+**Benefits:**
+- **Retry-friendly**: If normalize fails, re-run product_refresh without re-crawling
+- **Replay-able**: Run product_refresh against any historical payload
+- **Faster refreshes**: Local file read vs network call
+- **Historical diffs**: Compare payloads to see what changed between crawls
+- **Less API pressure**: Only payload_fetch hits Dutchie
+
+---
+
+## Task Chaining
+
+Tasks automatically queue follow-up tasks upon successful completion. This creates two main flows:
+
+### Discovery Flow (New Stores)
+
+When `store_discovery` finds new dispensaries, they automatically get their initial product data:
+
+```
+store_discovery
+  └─► Discovers new locations via Dutchie GraphQL
+  └─► Auto-promotes valid locations to dispensaries table
+  └─► Collects newDispensaryIds[] from promotions
+  └─► Returns { newStoreIds: [...] } in result
+
+chainNextTask() detects newStoreIds
+  └─► Creates product_discovery task for each new store
+
+product_discovery
+  └─► Calls handlePayloadFetch() internally
+  └─► payload_fetch hits Dutchie API
+  └─► Saves raw JSON to /storage/payloads/
+  └─► Queues product_refresh task with payload_id
+
+product_refresh
+  └─► Loads payload from filesystem
+  └─► Normalizes and upserts to store_products
+  └─► Creates snapshots, downloads images
+```
+
+**Complete Discovery Chain:**
+```
+store_discovery → product_discovery → payload_fetch → product_refresh
+                        (internal call)    (queues next)
+```
+
+### Scheduled Flow (Existing Stores)
+
+For existing stores, `payload_fetch_all` schedule runs every 4 hours:
+
+```
+TaskScheduler (every 60s)
+  └─► Checks task_schedules for due schedules
+  └─► payload_fetch_all is due
+  └─► Generates payload_fetch task for each dispensary
+
+payload_fetch
+  └─► Hits Dutchie GraphQL API
+  └─► Saves raw JSON to /storage/payloads/
+  └─► Queues product_refresh task with payload_id
+
+product_refresh
+  └─► Loads payload from filesystem (NOT API)
+  └─► Normalizes via DutchieNormalizer
+  └─► Upserts to store_products
+  └─► Creates snapshots
+```
+
+**Complete Scheduled Chain:**
+```
+payload_fetch → product_refresh
+  (queues)        (reads local)
+```
+
+### Chaining Implementation
+
+Task chaining is handled in two places:
+
+1. **Internal chaining (handler calls handler):**
+   - `product_discovery` calls `handlePayloadFetch()` directly
+
+2. **External chaining (chainNextTask() in task-service.ts):**
+   - Called after task completion
+   - `store_discovery` → queues `product_discovery` for each newStoreId
+
+3. **Queue-based chaining (taskService.createTask):**
+   - `payload_fetch` queues `product_refresh` with `payload: { payload_id }`
+
+---
+
+## Payload API Endpoints
+
+Raw crawl payloads can be accessed via the Payloads API:
+
+| Endpoint | Method | Description |
+|----------|--------|-------------|
+| `GET /api/payloads` | GET | List payload metadata (paginated) |
+| `GET /api/payloads/:id` | GET | Get payload metadata by ID |
+| `GET /api/payloads/:id/data` | GET | Get full payload JSON (decompressed) |
+| `GET /api/payloads/store/:dispensaryId` | GET | List payloads for a store |
+| `GET /api/payloads/store/:dispensaryId/latest` | GET | Get latest payload for a store |
+| `GET /api/payloads/store/:dispensaryId/diff` | GET | Diff two payloads for changes |
+
+### Payload Diff Response
+
+The diff endpoint returns:
+```json
+{
+  "success": true,
+  "from": { "id": 123, "fetchedAt": "...", "productCount": 100 },
+  "to": { "id": 456, "fetchedAt": "...", "productCount": 105 },
+  "diff": {
+    "added": 10,
+    "removed": 5,
+    "priceChanges": 8,
+    "stockChanges": 12
+  },
+  "details": {
+    "added": [...],
+    "removed": [...],
+    "priceChanges": [...],
+    "stockChanges": [...]
+  }
+}
+```
+
+---
+
+## API Endpoints
+
+### Schedules (NEW)
+
+| Endpoint | Method | Description |
+|----------|--------|-------------|
+| `GET /api/schedules` | GET | List all schedules |
+| `PUT /api/schedules/:id` | PUT | Update schedule |
+| `POST /api/schedules/:id/trigger` | POST | Run schedule immediately |
+
+### Task Creation (rewired 2024-12-10)
+
+| Endpoint | Method | Description |
+|----------|--------|-------------|
+| `POST /api/job-queue/enqueue` | POST | Create single task |
+| `POST /api/job-queue/enqueue-batch` | POST | Create batch tasks |
+| `POST /api/job-queue/enqueue-state` | POST | Create tasks for state |
+| `POST /api/tasks` | POST | Direct task creation |
+
+### Task Management
+
+| Endpoint | Method | Description |
+|----------|--------|-------------|
+| `GET /api/tasks` | GET | List tasks |
+| `GET /api/tasks/:id` | GET | Get single task |
+| `GET /api/tasks/counts` | GET | Task counts by status |
+| `POST /api/tasks/recover-stale` | POST | Recover stale tasks |
+
+---
+
+## Key Files
+
+| File | Purpose |
+|------|---------|
+| `src/services/task-scheduler.ts` | **NEW** - DB-driven scheduler |
+| `src/tasks/task-worker.ts` | Worker that processes tasks |
+| `src/tasks/task-service.ts` | Task CRUD operations |
+| `src/tasks/handlers/payload-fetch.ts` | **NEW** - Fetches from API, saves to disk |
+| `src/tasks/handlers/product-refresh.ts` | **CHANGED** - Reads from disk, processes to DB |
+| `src/utils/payload-storage.ts` | **NEW** - Payload save/load utilities |
+| `src/routes/tasks.ts` | Task API endpoints |
+| `src/routes/job-queue.ts` | Job Queue UI endpoints (rewired) |
+| `migrations/079_task_schedules.sql` | Schedule table |
+| `migrations/080_raw_crawl_payloads.sql` | Payload metadata table |
+| `migrations/081_payload_fetch_columns.sql` | payload, last_fetch_at columns |
+| `migrations/074_worker_task_queue.sql` | Task queue table |
+
+---
+
+## Legacy Code (DEPRECATED)
+
+| File | Status | Replacement |
+|------|--------|-------------|
+| `src/services/scheduler.ts` | DEPRECATED | `task-scheduler.ts` |
+| `dispensary_crawl_jobs` table | ORPHANED | `worker_tasks` |
+| `job_schedules` table | LEGACY | `task_schedules` |
+
+---
+
+## Dashboard Integration
+
+Both pages remain wired to the dashboard:
+
+| Page | Data Source | Actions |
+|------|-------------|---------|
+| **Job Queue** | `worker_tasks`, `task_schedules` | Create tasks, view schedules |
+| **Task Queue** | `worker_tasks` | View tasks, recover stale |
+
+---
+
+## Multi-Replica Safety
+
+The scheduler uses `SELECT FOR UPDATE SKIP LOCKED` to ensure:
+
+1. **Only one replica** executes a schedule at a time
+2. **No duplicate tasks** created
+3. **Survives pod restarts** - state in DB, not memory
+4. **Self-healing** - recovers stale tasks on startup
+
+```sql
+-- This query is atomic across all API server replicas
+SELECT * FROM task_schedules
+WHERE enabled = true AND next_run_at <= NOW()
+FOR UPDATE SKIP LOCKED
+```
+
+---
+
+## Worker Scaling (K8s)
+
+Workers run as a StatefulSet in Kubernetes. You can scale from the admin UI or CLI.
+
+### From Admin UI
+
+The Workers page (`/admin/workers`) provides:
+- Current replica count display
+- Scale up/down buttons
+- Target replica input
+
+### API Endpoints
+
+| Endpoint | Method | Description |
+|----------|--------|-------------|
+| `GET /api/workers/k8s/replicas` | GET | Get current/desired replica counts |
+| `POST /api/workers/k8s/scale` | POST | Scale to N replicas (body: `{ replicas: N }`) |
+
+### From CLI
+
+```bash
+# View current replicas
+kubectl get statefulset scraper-worker -n dispensary-scraper
+
+# Scale to 10 workers
+kubectl scale statefulset scraper-worker -n dispensary-scraper --replicas=10
+
+# Scale down to 3 workers
+kubectl scale statefulset scraper-worker -n dispensary-scraper --replicas=3
+```
+
+### Configuration
+
+Environment variables for the API server:
+
+| Variable | Default | Description |
+|----------|---------|-------------|
+| `K8S_NAMESPACE` | `dispensary-scraper` | Kubernetes namespace |
+| `K8S_WORKER_STATEFULSET` | `scraper-worker` | StatefulSet name |
+
+### RBAC Requirements
+
+The API server pod needs these K8s permissions:
+
+```yaml
+apiVersion: rbac.authorization.k8s.io/v1
+kind: Role
+metadata:
+  name: worker-scaler
+  namespace: dispensary-scraper
+rules:
+- apiGroups: ["apps"]
+  resources: ["statefulsets"]
+  verbs: ["get", "patch"]
+---
+apiVersion: rbac.authorization.k8s.io/v1
+kind: RoleBinding
+metadata:
+  name: scraper-worker-scaler
+  namespace: dispensary-scraper
+subjects:
+- kind: ServiceAccount
+  name: default
+  namespace: dispensary-scraper
+roleRef:
+  kind: Role
+  name: worker-scaler
+  apiGroup: rbac.authorization.k8s.io
+```

backend/docs/_archive/WORKER_TASK_ARCHITECTURE.md

@@ -0,0 +1,639 @@
+# Worker Task Architecture
+
+This document describes the unified task-based worker system that replaces the legacy fragmented job systems.
+
+## Overview
+
+The task worker architecture provides a single, unified system for managing all background work in CannaiQ:
+
+- **Store discovery** - Find new dispensaries on platforms
+- **Entry point discovery** - Resolve platform IDs from menu URLs
+- **Product discovery** - Initial product fetch for new stores
+- **Product resync** - Regular price/stock updates for existing stores
+- **Analytics refresh** - Refresh materialized views and analytics
+
+## Architecture
+
+### Database Tables
+
+**`worker_tasks`** - Central task queue
+```sql
+CREATE TABLE worker_tasks (
+  id SERIAL PRIMARY KEY,
+  role task_role NOT NULL,           -- What type of work
+  dispensary_id INTEGER,              -- Which store (if applicable)
+  platform VARCHAR(50),               -- Which platform (dutchie, etc.)
+  status task_status DEFAULT 'pending',
+  priority INTEGER DEFAULT 0,         -- Higher = process first
+  scheduled_for TIMESTAMP,            -- Don't process before this time
+  worker_id VARCHAR(100),             -- Which worker claimed it
+  claimed_at TIMESTAMP,
+  started_at TIMESTAMP,
+  completed_at TIMESTAMP,
+  last_heartbeat_at TIMESTAMP,        -- For stale detection
+  result JSONB,                       -- Output from handler
+  error_message TEXT,
+  retry_count INTEGER DEFAULT 0,
+  max_retries INTEGER DEFAULT 3,
+  created_at TIMESTAMP DEFAULT NOW(),
+  updated_at TIMESTAMP DEFAULT NOW()
+);
+```
+
+**Key indexes:**
+- `idx_worker_tasks_pending_priority` - For efficient task claiming
+- `idx_worker_tasks_active_dispensary` - Prevents concurrent tasks per store (partial unique index)
+
+### Task Roles
+
+| Role | Purpose | Per-Store | Scheduled |
+|------|---------|-----------|-----------|
+| `store_discovery` | Find new stores on a platform | No | Daily |
+| `entry_point_discovery` | Resolve platform IDs | Yes | On-demand |
+| `product_discovery` | Initial product fetch | Yes | After entry_point |
+| `product_resync` | Price/stock updates | Yes | Every 4 hours |
+| `analytics_refresh` | Refresh MVs | No | Daily |
+
+### Task Lifecycle
+
+```
+pending → claimed → running → completed
+                  ↓
+                failed
+```
+
+1. **pending** - Task is waiting to be picked up
+2. **claimed** - Worker has claimed it (atomic via SELECT FOR UPDATE SKIP LOCKED)
+3. **running** - Worker is actively processing
+4. **completed** - Task finished successfully
+5. **failed** - Task encountered an error
+6. **stale** - Task lost its worker (recovered automatically)
+
+## Files
+
+### Core Files
+
+| File | Purpose |
+|------|---------|
+| `src/tasks/task-service.ts` | TaskService - CRUD, claiming, capacity metrics |
+| `src/tasks/task-worker.ts` | TaskWorker - Main worker loop |
+| `src/tasks/index.ts` | Module exports |
+| `src/routes/tasks.ts` | API endpoints |
+| `migrations/074_worker_task_queue.sql` | Database schema |
+
+### Task Handlers
+
+| File | Role |
+|------|------|
+| `src/tasks/handlers/store-discovery.ts` | `store_discovery` |
+| `src/tasks/handlers/entry-point-discovery.ts` | `entry_point_discovery` |
+| `src/tasks/handlers/product-discovery.ts` | `product_discovery` |
+| `src/tasks/handlers/product-resync.ts` | `product_resync` |
+| `src/tasks/handlers/analytics-refresh.ts` | `analytics_refresh` |
+
+## Running Workers
+
+### Environment Variables
+
+| Variable | Default | Description |
+|----------|---------|-------------|
+| `WORKER_ROLE` | (required) | Which task role to process |
+| `WORKER_ID` | auto-generated | Custom worker identifier |
+| `POLL_INTERVAL_MS` | 5000 | How often to check for tasks |
+| `HEARTBEAT_INTERVAL_MS` | 30000 | How often to update heartbeat |
+
+### Starting a Worker
+
+```bash
+# Start a product resync worker
+WORKER_ROLE=product_resync npx tsx src/tasks/task-worker.ts
+
+# Start with custom ID
+WORKER_ROLE=product_resync WORKER_ID=resync-1 npx tsx src/tasks/task-worker.ts
+
+# Start multiple workers for different roles
+WORKER_ROLE=store_discovery npx tsx src/tasks/task-worker.ts &
+WORKER_ROLE=product_resync npx tsx src/tasks/task-worker.ts &
+```
+
+### Kubernetes Deployment
+
+```yaml
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: task-worker-resync
+spec:
+  replicas: 3
+  template:
+    spec:
+      containers:
+      - name: worker
+        image: code.cannabrands.app/creationshop/dispensary-scraper:latest
+        command: ["npx", "tsx", "src/tasks/task-worker.ts"]
+        env:
+        - name: WORKER_ROLE
+          value: "product_resync"
+```
+
+## API Endpoints
+
+### Task Management
+
+| Endpoint | Method | Description |
+|----------|--------|-------------|
+| `/api/tasks` | GET | List tasks with filters |
+| `/api/tasks` | POST | Create a new task |
+| `/api/tasks/:id` | GET | Get task by ID |
+| `/api/tasks/counts` | GET | Get counts by status |
+| `/api/tasks/capacity` | GET | Get capacity metrics |
+| `/api/tasks/capacity/:role` | GET | Get role-specific capacity |
+| `/api/tasks/recover-stale` | POST | Recover tasks from dead workers |
+
+### Task Generation
+
+| Endpoint | Method | Description |
+|----------|--------|-------------|
+| `/api/tasks/generate/resync` | POST | Generate daily resync tasks |
+| `/api/tasks/generate/discovery` | POST | Create store discovery task |
+
+### Migration (from legacy systems)
+
+| Endpoint | Method | Description |
+|----------|--------|-------------|
+| `/api/tasks/migration/status` | GET | Compare old vs new systems |
+| `/api/tasks/migration/disable-old-schedules` | POST | Disable job_schedules |
+| `/api/tasks/migration/cancel-pending-crawl-jobs` | POST | Cancel old crawl jobs |
+| `/api/tasks/migration/create-resync-tasks` | POST | Create tasks for all stores |
+| `/api/tasks/migration/full-migrate` | POST | One-click migration |
+
+### Role-Specific Endpoints
+
+| Endpoint | Method | Description |
+|----------|--------|-------------|
+| `/api/tasks/role/:role/last-completion` | GET | Last completion time |
+| `/api/tasks/role/:role/recent` | GET | Recent completions |
+| `/api/tasks/store/:id/active` | GET | Check if store has active task |
+
+## Capacity Planning
+
+The `v_worker_capacity` view provides real-time metrics:
+
+```sql
+SELECT * FROM v_worker_capacity;
+```
+
+Returns:
+- `pending_tasks` - Tasks waiting to be claimed
+- `ready_tasks` - Tasks ready now (scheduled_for is null or past)
+- `claimed_tasks` - Tasks claimed but not started
+- `running_tasks` - Tasks actively processing
+- `completed_last_hour` - Recent completions
+- `failed_last_hour` - Recent failures
+- `active_workers` - Workers with recent heartbeats
+- `avg_duration_sec` - Average task duration
+- `tasks_per_worker_hour` - Throughput estimate
+- `estimated_hours_to_drain` - Time to clear queue
+
+### Scaling Recommendations
+
+```javascript
+// API: GET /api/tasks/capacity/:role
+{
+  "role": "product_resync",
+  "pending_tasks": 500,
+  "active_workers": 3,
+  "workers_needed": {
+    "for_1_hour": 10,
+    "for_4_hours": 3,
+    "for_8_hours": 2
+  }
+}
+```
+
+## Task Chaining
+
+Tasks can automatically create follow-up tasks:
+
+```
+store_discovery → entry_point_discovery → product_discovery
+                              ↓
+                     (store has platform_dispensary_id)
+                              ↓
+                     Daily resync tasks
+```
+
+The `chainNextTask()` method handles this automatically.
+
+## Stale Task Recovery
+
+Tasks are considered stale if `last_heartbeat_at` is older than the threshold (default 10 minutes).
+
+```sql
+SELECT recover_stale_tasks(10); -- 10 minute threshold
+```
+
+Or via API:
+```bash
+curl -X POST /api/tasks/recover-stale \
+  -H 'Content-Type: application/json' \
+  -d '{"threshold_minutes": 10}'
+```
+
+## Migration from Legacy Systems
+
+### Legacy Systems Replaced
+
+1. **job_schedules + job_run_logs** - Scheduled job definitions
+2. **dispensary_crawl_jobs** - Per-dispensary crawl queue
+3. **SyncOrchestrator + HydrationWorker** - Raw payload processing
+
+### Migration Steps
+
+**Option 1: One-Click Migration**
+```bash
+curl -X POST /api/tasks/migration/full-migrate
+```
+
+This will:
+1. Disable all job_schedules
+2. Cancel pending dispensary_crawl_jobs
+3. Generate resync tasks for all stores
+4. Create discovery and analytics tasks
+
+**Option 2: Manual Migration**
+```bash
+# 1. Check current status
+curl /api/tasks/migration/status
+
+# 2. Disable old schedules
+curl -X POST /api/tasks/migration/disable-old-schedules
+
+# 3. Cancel pending crawl jobs
+curl -X POST /api/tasks/migration/cancel-pending-crawl-jobs
+
+# 4. Create resync tasks
+curl -X POST /api/tasks/migration/create-resync-tasks \
+  -H 'Content-Type: application/json' \
+  -d '{"state_code": "AZ"}'
+
+# 5. Generate daily resync schedule
+curl -X POST /api/tasks/generate/resync \
+  -H 'Content-Type: application/json' \
+  -d '{"batches_per_day": 6}'
+```
+
+## Per-Store Locking
+
+The system prevents concurrent tasks for the same store using a partial unique index:
+
+```sql
+CREATE UNIQUE INDEX idx_worker_tasks_active_dispensary
+ON worker_tasks (dispensary_id)
+WHERE dispensary_id IS NOT NULL
+AND status IN ('claimed', 'running');
+```
+
+This ensures only one task can be active per store at any time.
+
+## Task Priority
+
+Tasks are claimed in priority order (higher first), then by creation time:
+
+```sql
+ORDER BY priority DESC, created_at ASC
+```
+
+Default priorities:
+- `store_discovery`: 0
+- `entry_point_discovery`: 10 (high - new stores)
+- `product_discovery`: 10 (high - new stores)
+- `product_resync`: 0
+- `analytics_refresh`: 0
+
+## Scheduled Tasks
+
+Tasks can be scheduled for future execution:
+
+```javascript
+await taskService.createTask({
+  role: 'product_resync',
+  dispensary_id: 123,
+  scheduled_for: new Date('2025-01-10T06:00:00Z'),
+});
+```
+
+The `generate_resync_tasks()` function creates staggered tasks throughout the day:
+
+```sql
+SELECT generate_resync_tasks(6, '2025-01-10'); -- 6 batches = every 4 hours
+```
+
+## Dashboard Integration
+
+The admin dashboard shows task queue status in the main overview:
+
+```
+Task Queue Summary
+------------------
+Pending:   45
+Running:   3
+Completed: 1,234
+Failed:    12
+```
+
+Full task management is available at `/admin/tasks`.
+
+## Error Handling
+
+Failed tasks include the error message in `error_message` and can be retried:
+
+```sql
+-- View failed tasks
+SELECT id, role, dispensary_id, error_message, retry_count
+FROM worker_tasks
+WHERE status = 'failed'
+ORDER BY completed_at DESC
+LIMIT 20;
+
+-- Retry failed tasks
+UPDATE worker_tasks
+SET status = 'pending', retry_count = retry_count + 1
+WHERE status = 'failed' AND retry_count < max_retries;
+```
+
+## Concurrent Task Processing (Added 2024-12)
+
+Workers can now process multiple tasks concurrently within a single worker instance. This improves throughput by utilizing async I/O efficiently.
+
+### Architecture
+
+```
+┌─────────────────────────────────────────────────────────────┐
+│                         Pod (K8s)                           │
+│                                                             │
+│  ┌─────────────────────────────────────────────────────┐   │
+│  │                    TaskWorker                        │   │
+│  │                                                      │   │
+│  │  ┌─────────┐  ┌─────────┐  ┌─────────┐             │   │
+│  │  │ Task 1  │  │ Task 2  │  │ Task 3  │  (concurrent)│   │
+│  │  └─────────┘  └─────────┘  └─────────┘             │   │
+│  │                                                      │   │
+│  │  Resource Monitor                                    │   │
+│  │  ├── Memory: 65% (threshold: 85%)                   │   │
+│  │  ├── CPU: 45% (threshold: 90%)                      │   │
+│  │  └── Status: Normal                                  │   │
+│  └─────────────────────────────────────────────────────┘   │
+└─────────────────────────────────────────────────────────────┘
+```
+
+### Environment Variables
+
+| Variable | Default | Description |
+|----------|---------|-------------|
+| `MAX_CONCURRENT_TASKS` | 3 | Maximum tasks a worker will run concurrently |
+| `MEMORY_BACKOFF_THRESHOLD` | 0.85 | Back off when heap memory exceeds 85% |
+| `CPU_BACKOFF_THRESHOLD` | 0.90 | Back off when CPU exceeds 90% |
+| `BACKOFF_DURATION_MS` | 10000 | How long to wait when backing off (10s) |
+
+### How It Works
+
+1. **Main Loop**: Worker continuously tries to fill up to `MAX_CONCURRENT_TASKS`
+2. **Resource Monitoring**: Before claiming a new task, worker checks memory and CPU
+3. **Backoff**: If resources exceed thresholds, worker pauses and stops claiming new tasks
+4. **Concurrent Execution**: Tasks run in parallel using `Promise` - they don't block each other
+5. **Graceful Shutdown**: On SIGTERM/decommission, worker stops claiming but waits for active tasks
+
+### Resource Monitoring
+
+```typescript
+// ResourceStats interface
+interface ResourceStats {
+  memoryPercent: number;    // Current heap usage as decimal (0.0-1.0)
+  memoryMb: number;         // Current heap used in MB
+  memoryTotalMb: number;    // Total heap available in MB
+  cpuPercent: number;       // CPU usage as percentage (0-100)
+  isBackingOff: boolean;    // True if worker is in backoff state
+  backoffReason: string;    // Why the worker is backing off
+}
+```
+
+### Heartbeat Data
+
+Workers report the following in their heartbeat:
+
+```json
+{
+  "worker_id": "worker-abc123",
+  "current_task_id": 456,
+  "current_task_ids": [456, 457, 458],
+  "active_task_count": 3,
+  "max_concurrent_tasks": 3,
+  "status": "active",
+  "resources": {
+    "memory_mb": 256,
+    "memory_total_mb": 512,
+    "memory_rss_mb": 320,
+    "memory_percent": 50,
+    "cpu_user_ms": 12500,
+    "cpu_system_ms": 3200,
+    "cpu_percent": 45,
+    "is_backing_off": false,
+    "backoff_reason": null
+  }
+}
+```
+
+### Backoff Behavior
+
+When resources exceed thresholds:
+
+1. Worker logs the backoff reason:
+   ```
+   [TaskWorker] MyWorker backing off: Memory at 87.3% (threshold: 85%)
+   ```
+
+2. Worker stops claiming new tasks but continues existing tasks
+
+3. After `BACKOFF_DURATION_MS`, worker rechecks resources
+
+4. When resources return to normal:
+   ```
+   [TaskWorker] MyWorker resuming normal operation
+   ```
+
+### UI Display
+
+The Workers Dashboard shows:
+
+- **Tasks Column**: `2/3 tasks` (active/max concurrent)
+- **Resources Column**: Memory % and CPU % with color coding
+  - Green: < 50%
+  - Yellow: 50-74%
+  - Amber: 75-89%
+  - Red: 90%+
+- **Backing Off**: Orange warning badge when worker is in backoff state
+
+### Task Count Badge Details
+
+```
+┌─────────────────────────────────────────────┐
+│ Worker: "MyWorker"                          │
+│ Tasks: 2/3 tasks  #456, #457                │
+│ Resources: 🧠 65%  💻 45%                    │
+│ Status: ● Active                            │
+└─────────────────────────────────────────────┘
+```
+
+### Best Practices
+
+1. **Start Conservative**: Use `MAX_CONCURRENT_TASKS=3` initially
+2. **Monitor Resources**: Watch for frequent backoffs in logs
+3. **Tune Per Workload**: I/O-bound tasks benefit from higher concurrency
+4. **Scale Horizontally**: Add more pods rather than cranking concurrency too high
+
+### Code References
+
+| File | Purpose |
+|------|---------|
+| `src/tasks/task-worker.ts:68-71` | Concurrency environment variables |
+| `src/tasks/task-worker.ts:104-111` | ResourceStats interface |
+| `src/tasks/task-worker.ts:149-179` | getResourceStats() method |
+| `src/tasks/task-worker.ts:184-196` | shouldBackOff() method |
+| `src/tasks/task-worker.ts:462-516` | mainLoop() with concurrent claiming |
+| `src/routes/worker-registry.ts:148-195` | Heartbeat endpoint handling |
+| `cannaiq/src/pages/WorkersDashboard.tsx:233-305` | UI components for resources |
+
+## Browser Task Memory Limits (Updated 2025-12)
+
+Browser-based tasks (Puppeteer/Chrome) have strict memory constraints that limit concurrency.
+
+### Why Browser Tasks Are Different
+
+Each browser task launches a Chrome process. Unlike I/O-bound API calls, browsers consume significant RAM:
+
+| Component | RAM Usage |
+|-----------|-----------|
+| Node.js runtime | ~150 MB |
+| Chrome browser (base) | ~200-250 MB |
+| Dutchie menu page (loaded) | ~100-150 MB |
+| **Per browser total** | **~350-450 MB** |
+
+### Memory Math for Pod Limits
+
+```
+Pod memory limit:     2 GB (2000 MB)
+Node.js runtime:      -150 MB
+Safety buffer:        -100 MB
+────────────────────────────────
+Available for browsers: 1750 MB
+
+Per browser + page:    ~400 MB
+
+Max browsers: 1750 ÷ 400 = ~4 browsers
+
+Recommended: 3 browsers (leaves headroom for spikes)
+```
+
+### MAX_CONCURRENT_TASKS for Browser Tasks
+
+| Browsers per Pod | RAM Used | Risk Level |
+|------------------|----------|------------|
+| 1 | ~500 MB | Very safe |
+| 2 | ~900 MB | Safe |
+| **3** | **~1.3 GB** | **Recommended** |
+| 4 | ~1.7 GB | Tight (may OOM) |
+| 5+ | >2 GB | Will OOM crash |
+
+**CRITICAL**: `MAX_CONCURRENT_TASKS=3` is the maximum safe value for browser tasks with current pod limits.
+
+### Scaling Strategy
+
+Scale **horizontally** (more pods) rather than vertically (more concurrency per pod):
+
+```
+┌─────────────────────────────────────────────────────────────────────────┐
+│ Cluster: 8 pods × 3 browsers = 24 concurrent tasks                       │
+│                                                                          │
+│  ┌─────────────┐ ┌─────────────┐ ┌─────────────┐ ┌─────────────┐       │
+│  │ Pod 0       │ │ Pod 1       │ │ Pod 2       │ │ Pod 3       │       │
+│  │ 3 browsers  │ │ 3 browsers  │ │ 3 browsers  │ │ 3 browsers  │       │
+│  └─────────────┘ └─────────────┘ └─────────────┘ └─────────────┘       │
+│                                                                          │
+│  ┌─────────────┐ ┌─────────────┐ ┌─────────────┐ ┌─────────────┐       │
+│  │ Pod 4       │ │ Pod 5       │ │ Pod 6       │ │ Pod 7       │       │
+│  │ 3 browsers  │ │ 3 browsers  │ │ 3 browsers  │ │ 3 browsers  │       │
+│  └─────────────┘ └─────────────┘ └─────────────┘ └─────────────┘       │
+└─────────────────────────────────────────────────────────────────────────┘
+```
+
+### Browser Lifecycle Per Task
+
+Each task gets a fresh browser with fresh IP/identity:
+
+```
+1. Claim task from queue
+2. Get fresh proxy from pool
+3. Launch browser with proxy
+4. Run preflight (verify IP)
+5. Execute scrape
+6. Close browser
+7. Repeat
+```
+
+This ensures:
+- Fresh IP per task (proxy rotation)
+- Fresh fingerprint per task (UA rotation)
+- No cookie/session bleed between tasks
+- Predictable memory usage
+
+### Increasing Capacity
+
+To handle more concurrent tasks:
+
+1. **Add more pods** (up to 8 per CLAUDE.md limit)
+2. **Increase pod memory** (allows 4 browsers per pod):
+   ```yaml
+   resources:
+     limits:
+       memory: "2.5Gi"  # from 2Gi
+   ```
+
+**DO NOT** simply increase `MAX_CONCURRENT_TASKS` without also increasing pod memory limits.
+
+## Monitoring
+
+### Logs
+
+Workers log to stdout:
+```
+[TaskWorker] Starting worker worker-product_resync-a1b2c3d4 for role: product_resync
+[TaskWorker] Claimed task 123 (product_resync) for dispensary 456
+[TaskWorker] Task 123 completed successfully
+```
+
+### Health Check
+
+Check if workers are active:
+```sql
+SELECT worker_id, role, COUNT(*), MAX(last_heartbeat_at)
+FROM worker_tasks
+WHERE last_heartbeat_at > NOW() - INTERVAL '5 minutes'
+GROUP BY worker_id, role;
+```
+
+### Metrics
+
+```sql
+-- Tasks by status
+SELECT status, COUNT(*) FROM worker_tasks GROUP BY status;
+
+-- Tasks by role
+SELECT role, status, COUNT(*) FROM worker_tasks GROUP BY role, status;
+
+-- Average duration by role
+SELECT role, AVG(EXTRACT(EPOCH FROM (completed_at - started_at))) as avg_seconds
+FROM worker_tasks
+WHERE status = 'completed' AND completed_at > NOW() - INTERVAL '24 hours'
+GROUP BY role;
+```

backend/k8s/cronjob-ip2location.yaml

@@ -0,0 +1,69 @@
+apiVersion: batch/v1
+kind: CronJob
+metadata:
+  name: ip2location-update
+  namespace: default
+spec:
+  # Run on the 1st of every month at 3am UTC
+  schedule: "0 3 1 * *"
+  concurrencyPolicy: Forbid
+  successfulJobsHistoryLimit: 3
+  failedJobsHistoryLimit: 3
+  jobTemplate:
+    spec:
+      template:
+        spec:
+          containers:
+          - name: ip2location-updater
+            image: curlimages/curl:latest
+            command:
+            - /bin/sh
+            - -c
+            - |
+              set -e
+              echo "Downloading IP2Location LITE DB5..."
+
+              # Download to temp
+              cd /tmp
+              curl -L -o ip2location.zip "https://www.ip2location.com/download/?token=${IP2LOCATION_TOKEN}&file=DB5LITEBIN"
+
+              # Extract
+              unzip -o ip2location.zip
+
+              # Find and copy the BIN file
+              BIN_FILE=$(ls *.BIN 2>/dev/null | head -1)
+              if [ -z "$BIN_FILE" ]; then
+                echo "ERROR: No BIN file found"
+                exit 1
+              fi
+
+              # Copy to shared volume
+              cp "$BIN_FILE" /data/IP2LOCATION-LITE-DB5.BIN
+
+              echo "Done! Database updated: /data/IP2LOCATION-LITE-DB5.BIN"
+            env:
+            - name: IP2LOCATION_TOKEN
+              valueFrom:
+                secretKeyRef:
+                  name: dutchie-backend-secret
+                  key: IP2LOCATION_TOKEN
+            volumeMounts:
+            - name: ip2location-data
+              mountPath: /data
+          restartPolicy: OnFailure
+          volumes:
+          - name: ip2location-data
+            persistentVolumeClaim:
+              claimName: ip2location-pvc
+---
+apiVersion: v1
+kind: PersistentVolumeClaim
+metadata:
+  name: ip2location-pvc
+  namespace: default
+spec:
+  accessModes:
+    - ReadWriteOnce
+  resources:
+    requests:
+      storage: 100Mi

backend/k8s/deployment.yaml

@@ -26,6 +26,12 @@ spec:
             name: dutchie-backend-config
         - secretRef:
             name: dutchie-backend-secret
+        env:
+        - name: IP2LOCATION_DB_PATH
+          value: /data/ip2location/IP2LOCATION-LITE-DB5.BIN
+        volumeMounts:
+        - name: ip2location-data
+          mountPath: /data/ip2location
         resources:
           requests:
             memory: "256Mi"
@@ -45,3 +51,7 @@ spec:
             port: 3010
           initialDelaySeconds: 5
           periodSeconds: 5
+      volumes:
+      - name: ip2location-data
+        persistentVolumeClaim:
+          claimName: ip2location-pvc

backend/k8s/scraper-worker-statefulset.yaml

@@ -0,0 +1,77 @@
+apiVersion: v1
+kind: Service
+metadata:
+  name: scraper-worker
+  namespace: cannaiq
+  labels:
+    app: scraper-worker
+spec:
+  clusterIP: None  # Headless service required for StatefulSet
+  selector:
+    app: scraper-worker
+  ports:
+  - port: 3010
+    name: http
+---
+apiVersion: apps/v1
+kind: StatefulSet
+metadata:
+  name: scraper-worker
+  namespace: cannaiq
+spec:
+  serviceName: scraper-worker
+  replicas: 8
+  podManagementPolicy: Parallel  # Start all pods at once
+  updateStrategy:
+    type: OnDelete  # Pods only update when manually deleted - no automatic restarts
+  selector:
+    matchLabels:
+      app: scraper-worker
+  template:
+    metadata:
+      labels:
+        app: scraper-worker
+    spec:
+      terminationGracePeriodSeconds: 60
+      imagePullSecrets:
+      - name: regcred
+      containers:
+      - name: worker
+        image: git.spdy.io/creationshop/cannaiq:latest
+        imagePullPolicy: Always
+        command: ["node"]
+        args: ["dist/tasks/task-worker.js"]
+        env:
+        - name: WORKER_MODE
+          value: "true"
+        - name: POD_NAME
+          valueFrom:
+            fieldRef:
+              fieldPath: metadata.name
+        - name: MAX_CONCURRENT_TASKS
+          value: "50"
+        - name: API_BASE_URL
+          value: http://scraper
+        - name: NODE_OPTIONS
+          value: --max-old-space-size=1500
+        envFrom:
+        - configMapRef:
+            name: scraper-config
+        - secretRef:
+            name: scraper-secrets
+        resources:
+          requests:
+            cpu: 100m
+            memory: 1Gi
+          limits:
+            cpu: 500m
+            memory: 2Gi
+        livenessProbe:
+          exec:
+            command:
+            - /bin/sh
+            - -c
+            - pgrep -f 'task-worker' > /dev/null
+          initialDelaySeconds: 10
+          periodSeconds: 30
+          failureThreshold: 3

backend/migrations/008_proxy_locations.sql

@@ -1,18 +1,18 @@
 -- Add location columns to proxies table
 ALTER TABLE proxies
-ADD COLUMN city VARCHAR(100),
+ADD COLUMN IF NOT EXISTS city VARCHAR(100),
-ADD COLUMN state VARCHAR(100),
+ADD COLUMN IF NOT EXISTS state VARCHAR(100),
-ADD COLUMN country VARCHAR(100),
+ADD COLUMN IF NOT EXISTS country VARCHAR(100),
-ADD COLUMN country_code VARCHAR(2),
+ADD COLUMN IF NOT EXISTS country_code VARCHAR(2),
-ADD COLUMN location_updated_at TIMESTAMP;
+ADD COLUMN IF NOT EXISTS location_updated_at TIMESTAMP;
 
 -- Add index for location-based queries
-CREATE INDEX idx_proxies_location ON proxies(country_code, state, city);
+CREATE INDEX IF NOT EXISTS idx_proxies_location ON proxies(country_code, state, city);
 
 -- Add the same to failed_proxies table
 ALTER TABLE failed_proxies
-ADD COLUMN city VARCHAR(100),
+ADD COLUMN IF NOT EXISTS city VARCHAR(100),
-ADD COLUMN state VARCHAR(100),
+ADD COLUMN IF NOT EXISTS state VARCHAR(100),
-ADD COLUMN country VARCHAR(100),
+ADD COLUMN IF NOT EXISTS country VARCHAR(100),
-ADD COLUMN country_code VARCHAR(2),
+ADD COLUMN IF NOT EXISTS country_code VARCHAR(2),
-ADD COLUMN location_updated_at TIMESTAMP;
+ADD COLUMN IF NOT EXISTS location_updated_at TIMESTAMP;

backend/migrations/011_create_dispensaries_table.sql

@@ -1,6 +1,6 @@
 -- Create dispensaries table as single source of truth
 -- This consolidates azdhs_list (official data) + stores (menu data) into one table
-CREATE TABLE dispensaries (
+CREATE TABLE IF NOT EXISTS dispensaries (
   -- Primary key
   id SERIAL PRIMARY KEY,
 
@@ -43,11 +43,11 @@ CREATE TABLE dispensaries (
 );
 
 -- Create indexes for common queries
-CREATE INDEX idx_dispensaries_city ON dispensaries(city);
+CREATE INDEX IF NOT EXISTS idx_dispensaries_city ON dispensaries(city);
-CREATE INDEX idx_dispensaries_state ON dispensaries(state);
+CREATE INDEX IF NOT EXISTS idx_dispensaries_state ON dispensaries(state);
-CREATE INDEX idx_dispensaries_slug ON dispensaries(slug);
+CREATE INDEX IF NOT EXISTS idx_dispensaries_slug ON dispensaries(slug);
-CREATE INDEX idx_dispensaries_azdhs_id ON dispensaries(azdhs_id);
+CREATE INDEX IF NOT EXISTS idx_dispensaries_azdhs_id ON dispensaries(azdhs_id);
-CREATE INDEX idx_dispensaries_menu_status ON dispensaries(menu_scrape_status);
+CREATE INDEX IF NOT EXISTS idx_dispensaries_menu_status ON dispensaries(menu_scrape_status);
 
 -- Create index for location-based queries
-CREATE INDEX idx_dispensaries_location ON dispensaries(latitude, longitude) WHERE latitude IS NOT NULL AND longitude IS NOT NULL;
+CREATE INDEX IF NOT EXISTS idx_dispensaries_location ON dispensaries(latitude, longitude) WHERE latitude IS NOT NULL AND longitude IS NOT NULL;

backend/migrations/012_create_dispensary_changes_table.sql

@@ -1,6 +1,6 @@
 -- Create dispensary_changes table for change approval workflow
 -- This protects against accidental data destruction by requiring manual review
-CREATE TABLE dispensary_changes (
+CREATE TABLE IF NOT EXISTS dispensary_changes (
   id SERIAL PRIMARY KEY,
   dispensary_id INTEGER NOT NULL REFERENCES dispensaries(id) ON DELETE CASCADE,
 
@@ -26,10 +26,10 @@ CREATE TABLE dispensary_changes (
 );
 
 -- Create indexes for common queries
-CREATE INDEX idx_dispensary_changes_status ON dispensary_changes(status);
+CREATE INDEX IF NOT EXISTS idx_dispensary_changes_status ON dispensary_changes(status);
-CREATE INDEX idx_dispensary_changes_dispensary_status ON dispensary_changes(dispensary_id, status);
+CREATE INDEX IF NOT EXISTS idx_dispensary_changes_dispensary_status ON dispensary_changes(dispensary_id, status);
-CREATE INDEX idx_dispensary_changes_created_at ON dispensary_changes(created_at DESC);
+CREATE INDEX IF NOT EXISTS idx_dispensary_changes_created_at ON dispensary_changes(created_at DESC);
-CREATE INDEX idx_dispensary_changes_requires_recrawl ON dispensary_changes(requires_recrawl) WHERE requires_recrawl = TRUE;
+CREATE INDEX IF NOT EXISTS idx_dispensary_changes_requires_recrawl ON dispensary_changes(requires_recrawl) WHERE requires_recrawl = TRUE;
 
 -- Create function to automatically set requires_recrawl for website/menu_url changes
 CREATE OR REPLACE FUNCTION set_requires_recrawl()
@@ -42,7 +42,8 @@ BEGIN
 END;
 $$ LANGUAGE plpgsql;
 
--- Create trigger to call the function
+-- Create trigger to call the function (drop first to make idempotent)
+DROP TRIGGER IF EXISTS trigger_set_requires_recrawl ON dispensary_changes;
 CREATE TRIGGER trigger_set_requires_recrawl
   BEFORE INSERT ON dispensary_changes
   FOR EACH ROW

backend/migrations/013_populate_dispensaries_from_azdhs.sql

@@ -1,6 +1,7 @@
 -- Populate dispensaries table from azdhs_list
 -- This migrates all 182 AZDHS records with their enriched Google Maps data
 -- For multi-location dispensaries with duplicate slugs, append city name to make unique
+-- IDEMPOTENT: Uses ON CONFLICT DO NOTHING to skip already-imported records
 
 WITH ranked_dispensaries AS (
   SELECT
@@ -78,9 +79,10 @@ SELECT
   created_at,
   updated_at
 FROM ranked_dispensaries
-ORDER BY id;
+ORDER BY id
+ON CONFLICT (azdhs_id) DO NOTHING;
 
--- Verify the migration
+-- Verify the migration (idempotent - just logs, doesn't fail)
 DO $$
 DECLARE
   source_count INTEGER;
@@ -89,9 +91,11 @@ BEGIN
   SELECT COUNT(*) INTO source_count FROM azdhs_list;
   SELECT COUNT(*) INTO dest_count FROM dispensaries;
 
-  RAISE NOTICE 'Migration complete: % records from azdhs_list → % records in dispensaries', source_count, dest_count;
+  RAISE NOTICE 'Migration status: % records in azdhs_list, % records in dispensaries', source_count, dest_count;
 
-  IF source_count != dest_count THEN
+  IF dest_count >= source_count THEN
-    RAISE EXCEPTION 'Record count mismatch! Expected %, got %', source_count, dest_count;
+    RAISE NOTICE 'OK: dispensaries table has expected records';
+  ELSE
+    RAISE WARNING 'dispensaries has fewer records than azdhs_list (% vs %)', dest_count, source_count;
   END IF;
 END $$;

backend/migrations/014_add_dispensary_fk_to_tables.sql

@@ -3,15 +3,15 @@
 
 -- Add dispensary_id to products table
 ALTER TABLE products
-  ADD COLUMN dispensary_id INTEGER REFERENCES dispensaries(id) ON DELETE CASCADE;
+  ADD COLUMN IF NOT EXISTS dispensary_id INTEGER REFERENCES dispensaries(id) ON DELETE CASCADE;
 
 -- Add dispensary_id to categories table
 ALTER TABLE categories
-  ADD COLUMN dispensary_id INTEGER REFERENCES dispensaries(id) ON DELETE CASCADE;
+  ADD COLUMN IF NOT EXISTS dispensary_id INTEGER REFERENCES dispensaries(id) ON DELETE CASCADE;
 
 -- Create indexes for the new foreign keys
-CREATE INDEX idx_products_dispensary_id ON products(dispensary_id);
+CREATE INDEX IF NOT EXISTS idx_products_dispensary_id ON products(dispensary_id);
-CREATE INDEX idx_categories_dispensary_id ON categories(dispensary_id);
+CREATE INDEX IF NOT EXISTS idx_categories_dispensary_id ON categories(dispensary_id);
 
 -- NOTE: We'll populate these FKs and migrate data from stores in a separate data migration
 -- For now, new scrapers should use dispensary_id, but old store_id still works

backend/migrations/051_worker_definitions.sql

@@ -0,0 +1,119 @@
+-- Migration 051: Worker Definitions
+-- Creates a dedicated workers table for named workers with roles and assignments
+
+-- Workers table - defines named workers with roles
+CREATE TABLE IF NOT EXISTS workers (
+  id SERIAL PRIMARY KEY,
+  name VARCHAR(100) NOT NULL UNIQUE,
+  role VARCHAR(100) NOT NULL,
+  description TEXT,
+  enabled BOOLEAN DEFAULT TRUE,
+
+  -- Schedule configuration (for dedicated crawl workers)
+  schedule_type VARCHAR(50) DEFAULT 'interval', -- 'interval', 'cron', 'manual'
+  interval_minutes INTEGER DEFAULT 240,
+  cron_expression VARCHAR(100), -- e.g., '0 */4 * * *'
+  jitter_minutes INTEGER DEFAULT 30,
+
+  -- Assignment scope
+  assignment_type VARCHAR(50) DEFAULT 'all', -- 'all', 'state', 'dispensary', 'chain'
+  assigned_state_codes TEXT[], -- e.g., ['AZ', 'CA']
+  assigned_dispensary_ids INTEGER[],
+  assigned_chain_ids INTEGER[],
+
+  -- Job configuration
+  job_type VARCHAR(50) NOT NULL DEFAULT 'dutchie_product_crawl',
+  job_config JSONB DEFAULT '{}',
+  priority INTEGER DEFAULT 0,
+  max_concurrent INTEGER DEFAULT 1,
+
+  -- Status tracking
+  status VARCHAR(50) DEFAULT 'idle', -- 'idle', 'running', 'paused', 'error'
+  last_run_at TIMESTAMPTZ,
+  last_status VARCHAR(50),
+  last_error TEXT,
+  last_duration_ms INTEGER,
+  next_run_at TIMESTAMPTZ,
+  current_job_id INTEGER,
+
+  -- Metrics
+  total_runs INTEGER DEFAULT 0,
+  successful_runs INTEGER DEFAULT 0,
+  failed_runs INTEGER DEFAULT 0,
+  avg_duration_ms INTEGER,
+
+  created_at TIMESTAMPTZ DEFAULT NOW(),
+  updated_at TIMESTAMPTZ DEFAULT NOW()
+);
+
+-- Worker run history
+CREATE TABLE IF NOT EXISTS worker_runs (
+  id SERIAL PRIMARY KEY,
+  worker_id INTEGER NOT NULL REFERENCES workers(id) ON DELETE CASCADE,
+  started_at TIMESTAMPTZ DEFAULT NOW(),
+  completed_at TIMESTAMPTZ,
+  status VARCHAR(50) DEFAULT 'running', -- 'running', 'success', 'error', 'cancelled'
+  duration_ms INTEGER,
+
+  -- What was processed
+  jobs_created INTEGER DEFAULT 0,
+  jobs_completed INTEGER DEFAULT 0,
+  jobs_failed INTEGER DEFAULT 0,
+  dispensaries_crawled INTEGER DEFAULT 0,
+  products_found INTEGER DEFAULT 0,
+
+  error_message TEXT,
+  metadata JSONB DEFAULT '{}',
+
+  created_at TIMESTAMPTZ DEFAULT NOW()
+);
+
+-- Index for efficient lookups
+CREATE INDEX IF NOT EXISTS idx_workers_enabled ON workers(enabled) WHERE enabled = TRUE;
+CREATE INDEX IF NOT EXISTS idx_workers_next_run ON workers(next_run_at) WHERE enabled = TRUE;
+CREATE INDEX IF NOT EXISTS idx_workers_status ON workers(status);
+CREATE INDEX IF NOT EXISTS idx_worker_runs_worker_id ON worker_runs(worker_id);
+CREATE INDEX IF NOT EXISTS idx_worker_runs_started_at ON worker_runs(started_at DESC);
+
+-- Add worker_id to dispensary_crawl_jobs if not exists
+DO $$
+BEGIN
+  IF NOT EXISTS (
+    SELECT 1 FROM information_schema.columns
+    WHERE table_name = 'dispensary_crawl_jobs' AND column_name = 'assigned_worker_id'
+  ) THEN
+    ALTER TABLE dispensary_crawl_jobs ADD COLUMN assigned_worker_id INTEGER REFERENCES workers(id);
+  END IF;
+END $$;
+
+-- Migrate existing job_schedules workers to new workers table
+INSERT INTO workers (name, role, description, enabled, interval_minutes, jitter_minutes, job_type, job_config, last_run_at, last_status, last_error, last_duration_ms, next_run_at)
+SELECT
+  worker_name,
+  worker_role,
+  description,
+  enabled,
+  base_interval_minutes,
+  jitter_minutes,
+  job_name,
+  job_config,
+  last_run_at,
+  last_status,
+  last_error_message,
+  last_duration_ms,
+  next_run_at
+FROM job_schedules
+WHERE worker_name IS NOT NULL
+ON CONFLICT (name) DO UPDATE SET
+  updated_at = NOW();
+
+-- Available worker roles (reference)
+COMMENT ON TABLE workers IS 'Named workers with specific roles and assignments. Roles include:
+- product_sync: Crawls products from dispensary menus
+- store_discovery: Discovers new dispensary locations
+- entry_point_finder: Detects menu providers and resolves platform IDs
+- analytics_refresh: Refreshes materialized views and analytics
+- price_monitor: Monitors price changes and triggers alerts
+- inventory_sync: Syncs inventory levels
+- image_processor: Downloads and processes product images
+- data_validator: Validates data integrity';

backend/migrations/052_seo_settings.sql

@@ -0,0 +1,49 @@
+-- Migration 052: SEO Settings Table
+-- Key/value store for SEO Orchestrator configuration
+
+CREATE TABLE IF NOT EXISTS seo_settings (
+  id SERIAL PRIMARY KEY,
+  key TEXT UNIQUE NOT NULL,
+  value JSONB NOT NULL,
+  created_at TIMESTAMP DEFAULT NOW(),
+  updated_at TIMESTAMP DEFAULT NOW()
+);
+
+-- Create index on key for fast lookups
+CREATE INDEX IF NOT EXISTS idx_seo_settings_key ON seo_settings(key);
+
+-- Seed with default settings
+INSERT INTO seo_settings (key, value) VALUES
+  -- Section 1: Global Content Generation Settings
+  ('primary_prompt_template', '"You are a cannabis industry content expert. Generate SEO-optimized content for {{page_type}} pages about {{subject}}. Focus on: {{focus_areas}}. Maintain a {{tone}} tone and keep content {{length}}."'),
+  ('regeneration_prompt_template', '"Regenerate the following SEO content with fresh perspectives. Original topic: {{subject}}. Improve upon: {{improvement_areas}}. Maintain compliance with cannabis industry standards."'),
+  ('default_content_length', '"medium"'),
+  ('tone_voice', '"informational"'),
+
+  -- Section 2: Automatic Refresh Rules
+  ('auto_refresh_interval', '"weekly"'),
+  ('trigger_pct_product_change', 'true'),
+  ('trigger_pct_brand_change', 'true'),
+  ('trigger_new_stores', 'true'),
+  ('trigger_market_shift', 'false'),
+  ('webhook_url', '""'),
+  ('notify_on_trigger', 'false'),
+
+  -- Section 3: Page-Level Defaults
+  ('default_title_template', '"{{state_name}} Dispensaries | Find Cannabis Near You | CannaiQ"'),
+  ('default_meta_description_template', '"Discover the best dispensaries in {{state_name}}. Browse {{dispensary_count}}+ licensed retailers, compare prices, and find cannabis products near you."'),
+  ('default_slug_template', '"dispensaries-{{state_code_lower}}"'),
+  ('default_og_image_template', '"/images/seo/og-{{state_code_lower}}.jpg"'),
+  ('enable_ai_images', 'false'),
+
+  -- Section 4: Crawl / Dataset Configuration
+  ('primary_data_provider', '"cannaiq"'),
+  ('fallback_data_provider', '"dutchie"'),
+  ('min_data_freshness_hours', '24'),
+  ('stale_data_behavior', '"allow_with_warning"')
+ON CONFLICT (key) DO NOTHING;
+
+-- Record migration
+INSERT INTO schema_migrations (version, name, applied_at)
+VALUES ('052', 'seo_settings', NOW())
+ON CONFLICT (version) DO NOTHING;

backend/migrations/057_crawl_enabled_and_dutchie_verified.sql

@@ -0,0 +1,42 @@
+-- Migration 057: Add crawl_enabled and dutchie_verified fields to dispensaries
+--
+-- Purpose:
+--   1. Add crawl_enabled to control which dispensaries get crawled
+--   2. Add dutchie_verified to track Dutchie source-of-truth verification
+--   3. Default existing records to crawl_enabled = TRUE to preserve behavior
+--
+-- After this migration, run the harmonization script to:
+--   - Match dispensaries to Dutchie discoveries
+--   - Update platform_dispensary_id from Dutchie
+--   - Set dutchie_verified = TRUE for matches
+--   - Set crawl_enabled = FALSE for unverified records
+
+-- Add crawl_enabled column (defaults to true to not break existing crawls)
+ALTER TABLE dispensaries
+ADD COLUMN IF NOT EXISTS crawl_enabled BOOLEAN DEFAULT TRUE;
+
+-- Add dutchie_verified column to track if record is verified against Dutchie
+ALTER TABLE dispensaries
+ADD COLUMN IF NOT EXISTS dutchie_verified BOOLEAN DEFAULT FALSE;
+
+-- Add dutchie_verified_at timestamp
+ALTER TABLE dispensaries
+ADD COLUMN IF NOT EXISTS dutchie_verified_at TIMESTAMP WITH TIME ZONE;
+
+-- Add dutchie_discovery_id to link back to the discovery record
+ALTER TABLE dispensaries
+ADD COLUMN IF NOT EXISTS dutchie_discovery_id BIGINT REFERENCES dutchie_discovery_locations(id);
+
+-- Create index for crawl queries (only crawl enabled dispensaries)
+CREATE INDEX IF NOT EXISTS idx_dispensaries_crawl_enabled
+ON dispensaries(crawl_enabled, state)
+WHERE crawl_enabled = TRUE;
+
+-- Create index for dutchie verification status
+CREATE INDEX IF NOT EXISTS idx_dispensaries_dutchie_verified
+ON dispensaries(dutchie_verified, state);
+
+COMMENT ON COLUMN dispensaries.crawl_enabled IS 'Whether this dispensary should be included in crawl jobs. Set to FALSE for unverified or problematic records.';
+COMMENT ON COLUMN dispensaries.dutchie_verified IS 'Whether this dispensary has been verified against Dutchie source of truth (matched by slug or manually linked).';
+COMMENT ON COLUMN dispensaries.dutchie_verified_at IS 'Timestamp when Dutchie verification was completed.';
+COMMENT ON COLUMN dispensaries.dutchie_discovery_id IS 'Link to the dutchie_discovery_locations record this was matched/verified against.';

backend/migrations/065_slug_verification_tracking.sql

@@ -0,0 +1,56 @@
+-- Migration 065: Slug verification and data source tracking
+-- Adds columns to track when slug/menu data was verified and from what source
+
+-- Add slug verification columns to dispensaries
+ALTER TABLE dispensaries
+ADD COLUMN IF NOT EXISTS slug_source VARCHAR(50),
+ADD COLUMN IF NOT EXISTS slug_verified_at TIMESTAMPTZ,
+ADD COLUMN IF NOT EXISTS slug_status VARCHAR(20) DEFAULT 'unverified',
+ADD COLUMN IF NOT EXISTS menu_url_source VARCHAR(50),
+ADD COLUMN IF NOT EXISTS menu_url_verified_at TIMESTAMPTZ,
+ADD COLUMN IF NOT EXISTS platform_id_source VARCHAR(50),
+ADD COLUMN IF NOT EXISTS platform_id_verified_at TIMESTAMPTZ,
+ADD COLUMN IF NOT EXISTS country VARCHAR(2) DEFAULT 'US';
+
+-- Add index for finding unverified stores
+CREATE INDEX IF NOT EXISTS idx_dispensaries_slug_status
+ON dispensaries(slug_status)
+WHERE slug_status != 'verified';
+
+-- Add index for country
+CREATE INDEX IF NOT EXISTS idx_dispensaries_country
+ON dispensaries(country);
+
+-- Comment on columns
+COMMENT ON COLUMN dispensaries.slug_source IS 'Source of slug data: dutchie_api, manual, azdhs, discovery, etc.';
+COMMENT ON COLUMN dispensaries.slug_verified_at IS 'When the slug was last verified against the source';
+COMMENT ON COLUMN dispensaries.slug_status IS 'Status: unverified, verified, invalid, changed';
+COMMENT ON COLUMN dispensaries.menu_url_source IS 'Source of menu_url: dutchie_api, website_scrape, manual, etc.';
+COMMENT ON COLUMN dispensaries.menu_url_verified_at IS 'When the menu_url was last verified';
+COMMENT ON COLUMN dispensaries.platform_id_source IS 'Source of platform_dispensary_id: dutchie_api, graphql_resolution, etc.';
+COMMENT ON COLUMN dispensaries.platform_id_verified_at IS 'When the platform_dispensary_id was last verified';
+COMMENT ON COLUMN dispensaries.country IS 'ISO 2-letter country code: US, CA, etc.';
+
+-- Update Green Pharms Mesa with verified Dutchie data
+UPDATE dispensaries
+SET
+  slug = 'green-pharms-mesa',
+  menu_url = 'https://dutchie.com/embedded-menu/green-pharms-mesa',
+  menu_type = 'dutchie',
+  platform_dispensary_id = '68dc47a2af90f2e653f8df30',
+  slug_source = 'dutchie_api',
+  slug_verified_at = NOW(),
+  slug_status = 'verified',
+  menu_url_source = 'dutchie_api',
+  menu_url_verified_at = NOW(),
+  platform_id_source = 'dutchie_api',
+  platform_id_verified_at = NOW(),
+  updated_at = NOW()
+WHERE id = 232;
+
+-- Mark all other AZ dispensaries as needing verification
+UPDATE dispensaries
+SET slug_status = 'unverified'
+WHERE state = 'AZ'
+  AND id != 232
+  AND (slug_status IS NULL OR slug_status = 'unverified');

backend/migrations/066_dutchie_field_alignment.sql

@@ -0,0 +1,140 @@
+-- Migration 066: Align dispensaries and discovery_locations tables with Dutchie field names
+-- Uses snake_case convention (Postgres standard) mapped from Dutchie's camelCase
+--
+-- Changes:
+-- 1. dispensaries: rename address→address1, zip→zipcode, remove company_name
+-- 2. dispensaries: add missing Dutchie fields
+-- 3. dutchie_discovery_locations: add missing Dutchie fields
+
+-- ============================================================================
+-- DISPENSARIES TABLE
+-- ============================================================================
+
+-- Rename address to address1 (matches Dutchie's address1)
+ALTER TABLE dispensaries RENAME COLUMN address TO address1;
+
+-- Rename zip to zipcode (matches Dutchie's zip, but we use zipcode for clarity)
+ALTER TABLE dispensaries RENAME COLUMN zip TO zipcode;
+
+-- Drop company_name (redundant with name)
+ALTER TABLE dispensaries DROP COLUMN IF EXISTS company_name;
+
+-- Add address2
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS address2 VARCHAR(255);
+
+-- Add country
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS country VARCHAR(100) DEFAULT 'United States';
+
+-- Add timezone
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS timezone VARCHAR(50);
+
+-- Add email
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS email VARCHAR(255);
+
+-- Add description
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS description TEXT;
+
+-- Add logo_image (Dutchie: logoImage)
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS logo_image TEXT;
+
+-- Add banner_image (Dutchie: bannerImage)
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS banner_image TEXT;
+
+-- Add offer_pickup (Dutchie: offerPickup)
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS offer_pickup BOOLEAN DEFAULT TRUE;
+
+-- Add offer_delivery (Dutchie: offerDelivery)
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS offer_delivery BOOLEAN DEFAULT FALSE;
+
+-- Add offer_curbside_pickup (Dutchie: offerCurbsidePickup)
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS offer_curbside_pickup BOOLEAN DEFAULT FALSE;
+
+-- Add is_medical (Dutchie: isMedical)
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS is_medical BOOLEAN DEFAULT FALSE;
+
+-- Add is_recreational (Dutchie: isRecreational)
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS is_recreational BOOLEAN DEFAULT FALSE;
+
+-- Add chain_slug (Dutchie: chain)
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS chain_slug VARCHAR(255);
+
+-- Add enterprise_id (Dutchie: retailer.enterpriseId)
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS enterprise_id VARCHAR(100);
+
+-- Add status (Dutchie: status - open/closed)
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS status VARCHAR(50);
+
+-- Add c_name (Dutchie: cName - the URL slug used in embedded menus)
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS c_name VARCHAR(255);
+
+-- ============================================================================
+-- DUTCHIE_DISCOVERY_LOCATIONS TABLE
+-- ============================================================================
+
+-- Add phone
+ALTER TABLE dutchie_discovery_locations ADD COLUMN IF NOT EXISTS phone VARCHAR(50);
+
+-- Add website (Dutchie: embedBackUrl)
+ALTER TABLE dutchie_discovery_locations ADD COLUMN IF NOT EXISTS website TEXT;
+
+-- Add email
+ALTER TABLE dutchie_discovery_locations ADD COLUMN IF NOT EXISTS email VARCHAR(255);
+
+-- Add description
+ALTER TABLE dutchie_discovery_locations ADD COLUMN IF NOT EXISTS description TEXT;
+
+-- Add logo_image
+ALTER TABLE dutchie_discovery_locations ADD COLUMN IF NOT EXISTS logo_image TEXT;
+
+-- Add banner_image
+ALTER TABLE dutchie_discovery_locations ADD COLUMN IF NOT EXISTS banner_image TEXT;
+
+-- Add chain_slug
+ALTER TABLE dutchie_discovery_locations ADD COLUMN IF NOT EXISTS chain_slug VARCHAR(255);
+
+-- Add enterprise_id
+ALTER TABLE dutchie_discovery_locations ADD COLUMN IF NOT EXISTS enterprise_id VARCHAR(100);
+
+-- Add c_name
+ALTER TABLE dutchie_discovery_locations ADD COLUMN IF NOT EXISTS c_name VARCHAR(255);
+
+-- Add country
+ALTER TABLE dutchie_discovery_locations ADD COLUMN IF NOT EXISTS country VARCHAR(100) DEFAULT 'United States';
+
+-- Add store status
+ALTER TABLE dutchie_discovery_locations ADD COLUMN IF NOT EXISTS store_status VARCHAR(50);
+
+-- ============================================================================
+-- INDEXES
+-- ============================================================================
+
+-- Index for chain lookups
+CREATE INDEX IF NOT EXISTS idx_dispensaries_chain_slug ON dispensaries(chain_slug) WHERE chain_slug IS NOT NULL;
+CREATE INDEX IF NOT EXISTS idx_discovery_locations_chain_slug ON dutchie_discovery_locations(chain_slug) WHERE chain_slug IS NOT NULL;
+
+-- Index for enterprise lookups (for multi-location chains)
+CREATE INDEX IF NOT EXISTS idx_dispensaries_enterprise_id ON dispensaries(enterprise_id) WHERE enterprise_id IS NOT NULL;
+CREATE INDEX IF NOT EXISTS idx_discovery_locations_enterprise_id ON dutchie_discovery_locations(enterprise_id) WHERE enterprise_id IS NOT NULL;
+
+-- Index for c_name lookups
+CREATE INDEX IF NOT EXISTS idx_dispensaries_c_name ON dispensaries(c_name) WHERE c_name IS NOT NULL;
+
+-- ============================================================================
+-- COMMENTS
+-- ============================================================================
+
+COMMENT ON COLUMN dispensaries.address1 IS 'Street address line 1 (Dutchie: address1)';
+COMMENT ON COLUMN dispensaries.address2 IS 'Street address line 2 (Dutchie: address2)';
+COMMENT ON COLUMN dispensaries.zipcode IS 'ZIP/postal code (Dutchie: zip)';
+COMMENT ON COLUMN dispensaries.c_name IS 'Dutchie URL slug for embedded menus (Dutchie: cName)';
+COMMENT ON COLUMN dispensaries.chain_slug IS 'Chain identifier slug (Dutchie: chain)';
+COMMENT ON COLUMN dispensaries.enterprise_id IS 'Parent enterprise UUID (Dutchie: retailer.enterpriseId)';
+COMMENT ON COLUMN dispensaries.logo_image IS 'Logo image URL (Dutchie: logoImage)';
+COMMENT ON COLUMN dispensaries.banner_image IS 'Banner image URL (Dutchie: bannerImage)';
+COMMENT ON COLUMN dispensaries.offer_pickup IS 'Offers in-store pickup (Dutchie: offerPickup)';
+COMMENT ON COLUMN dispensaries.offer_delivery IS 'Offers delivery (Dutchie: offerDelivery)';
+COMMENT ON COLUMN dispensaries.offer_curbside_pickup IS 'Offers curbside pickup (Dutchie: offerCurbsidePickup)';
+COMMENT ON COLUMN dispensaries.is_medical IS 'Licensed for medical sales (Dutchie: isMedical)';
+COMMENT ON COLUMN dispensaries.is_recreational IS 'Licensed for recreational sales (Dutchie: isRecreational)';
+
+SELECT 'Migration 066 completed: Dutchie field alignment' as status;

backend/migrations/067_promotion_log.sql

@@ -0,0 +1,24 @@
+-- Promotion log table for tracking discovery → dispensary promotions
+-- Tracks validation and promotion actions for audit/review
+
+CREATE TABLE IF NOT EXISTS dutchie_promotion_log (
+  id SERIAL PRIMARY KEY,
+  discovery_id INTEGER REFERENCES dutchie_discovery_locations(id) ON DELETE SET NULL,
+  dispensary_id INTEGER REFERENCES dispensaries(id) ON DELETE SET NULL,
+  action VARCHAR(50) NOT NULL, -- 'validated', 'rejected', 'promoted_create', 'promoted_update', 'skipped'
+  state_code VARCHAR(10),
+  store_name VARCHAR(255),
+  validation_errors TEXT[], -- Array of error messages if rejected
+  field_changes JSONB, -- Before/after snapshot of changed fields
+  triggered_by VARCHAR(100) DEFAULT 'auto', -- 'auto', 'manual', 'api'
+  created_at TIMESTAMP WITH TIME ZONE DEFAULT CURRENT_TIMESTAMP
+);
+
+-- Indexes for efficient querying
+CREATE INDEX IF NOT EXISTS idx_promotion_log_discovery_id ON dutchie_promotion_log(discovery_id);
+CREATE INDEX IF NOT EXISTS idx_promotion_log_dispensary_id ON dutchie_promotion_log(dispensary_id);
+CREATE INDEX IF NOT EXISTS idx_promotion_log_action ON dutchie_promotion_log(action);
+CREATE INDEX IF NOT EXISTS idx_promotion_log_state_code ON dutchie_promotion_log(state_code);
+CREATE INDEX IF NOT EXISTS idx_promotion_log_created_at ON dutchie_promotion_log(created_at DESC);
+
+COMMENT ON TABLE dutchie_promotion_log IS 'Audit log for discovery location validation and promotion to dispensaries';

backend/migrations/068_crawler_status_alerts.sql

@@ -0,0 +1,95 @@
+-- Migration 068: Crawler Status Alerts
+-- Creates status_alerts table for dashboard notifications and status change logging
+
+-- ============================================================
+-- STATUS ALERTS TABLE
+-- ============================================================
+
+CREATE TABLE IF NOT EXISTS crawler_status_alerts (
+  id SERIAL PRIMARY KEY,
+
+  -- References
+  dispensary_id INTEGER REFERENCES dispensaries(id),
+  profile_id INTEGER REFERENCES dispensary_crawler_profiles(id),
+
+  -- Alert info
+  alert_type VARCHAR(50) NOT NULL, -- 'status_change', 'crawl_error', 'validation_failed', 'promoted', 'demoted'
+  severity VARCHAR(20) DEFAULT 'info', -- 'info', 'warning', 'error', 'critical'
+
+  -- Status transition
+  previous_status VARCHAR(50),
+  new_status VARCHAR(50),
+
+  -- Context
+  message TEXT,
+  error_details JSONB,
+  metadata JSONB, -- Additional context (product counts, error codes, etc.)
+
+  -- Tracking
+  acknowledged BOOLEAN DEFAULT FALSE,
+  acknowledged_at TIMESTAMP WITH TIME ZONE,
+  acknowledged_by VARCHAR(100),
+
+  -- Timestamps
+  created_at TIMESTAMP WITH TIME ZONE DEFAULT CURRENT_TIMESTAMP
+);
+
+-- Indexes for common queries
+CREATE INDEX IF NOT EXISTS idx_crawler_status_alerts_dispensary ON crawler_status_alerts(dispensary_id);
+CREATE INDEX IF NOT EXISTS idx_crawler_status_alerts_type ON crawler_status_alerts(alert_type);
+CREATE INDEX IF NOT EXISTS idx_crawler_status_alerts_severity ON crawler_status_alerts(severity);
+CREATE INDEX IF NOT EXISTS idx_crawler_status_alerts_unack ON crawler_status_alerts(acknowledged) WHERE acknowledged = FALSE;
+CREATE INDEX IF NOT EXISTS idx_crawler_status_alerts_created ON crawler_status_alerts(created_at DESC);
+
+-- ============================================================
+-- STATUS DEFINITIONS (for reference/validation)
+-- ============================================================
+
+COMMENT ON TABLE crawler_status_alerts IS 'Crawler status change notifications for dashboard alerting';
+COMMENT ON COLUMN crawler_status_alerts.alert_type IS 'Type: status_change, crawl_error, validation_failed, promoted, demoted';
+COMMENT ON COLUMN crawler_status_alerts.severity IS 'Severity: info, warning, error, critical';
+COMMENT ON COLUMN crawler_status_alerts.previous_status IS 'Previous crawler status before change';
+COMMENT ON COLUMN crawler_status_alerts.new_status IS 'New crawler status after change';
+
+-- ============================================================
+-- STATUS TRACKING ON PROFILES
+-- ============================================================
+
+-- Add columns for status tracking if not exists
+DO $$
+BEGIN
+  -- Consecutive success count for auto-promotion
+  IF NOT EXISTS (SELECT 1 FROM information_schema.columns
+    WHERE table_name = 'dispensary_crawler_profiles' AND column_name = 'consecutive_successes') THEN
+    ALTER TABLE dispensary_crawler_profiles ADD COLUMN consecutive_successes INTEGER DEFAULT 0;
+  END IF;
+
+  -- Consecutive failure count for auto-demotion
+  IF NOT EXISTS (SELECT 1 FROM information_schema.columns
+    WHERE table_name = 'dispensary_crawler_profiles' AND column_name = 'consecutive_failures') THEN
+    ALTER TABLE dispensary_crawler_profiles ADD COLUMN consecutive_failures INTEGER DEFAULT 0;
+  END IF;
+
+  -- Last status change timestamp
+  IF NOT EXISTS (SELECT 1 FROM information_schema.columns
+    WHERE table_name = 'dispensary_crawler_profiles' AND column_name = 'status_changed_at') THEN
+    ALTER TABLE dispensary_crawler_profiles ADD COLUMN status_changed_at TIMESTAMP WITH TIME ZONE;
+  END IF;
+
+  -- Status change reason
+  IF NOT EXISTS (SELECT 1 FROM information_schema.columns
+    WHERE table_name = 'dispensary_crawler_profiles' AND column_name = 'status_reason') THEN
+    ALTER TABLE dispensary_crawler_profiles ADD COLUMN status_reason TEXT;
+  END IF;
+END $$;
+
+-- ============================================================
+-- VALID STATUS VALUES
+-- ============================================================
+-- Status values for dispensary_crawler_profiles.status:
+--   'sandbox'      - Newly created, being validated
+--   'production'   - Healthy, actively crawled
+--   'needs_manual' - Requires human intervention
+--   'failing'      - Multiple consecutive failures
+--   'disabled'     - Manually disabled
+--   'legacy'       - No profile, uses default method (virtual status)

backend/migrations/069_six_stage_status.sql

@@ -0,0 +1,163 @@
+-- Migration 069: Seven-Stage Status System
+--
+-- Implements explicit 7-stage pipeline for store lifecycle:
+--   1. discovered - Found via Dutchie API, raw data
+--   2. validated  - Passed field checks, ready for promotion
+--   3. promoted   - In dispensaries table, has crawler profile
+--   4. sandbox    - First crawl attempted, testing
+--   5. hydrating  - Products are being loaded/updated
+--   6. production - Healthy, scheduled crawls via Horizon
+--   7. failing    - Crawl errors, needs attention
+
+-- ============================================================
+-- STAGE ENUM TYPE
+-- ============================================================
+
+DO $$
+BEGIN
+  -- Create enum if not exists
+  IF NOT EXISTS (SELECT 1 FROM pg_type WHERE typname = 'store_stage') THEN
+    CREATE TYPE store_stage AS ENUM (
+      'discovered',
+      'validated',
+      'promoted',
+      'sandbox',
+      'hydrating',
+      'production',
+      'failing'
+    );
+  END IF;
+END $$;
+
+-- ============================================================
+-- UPDATE DISCOVERY LOCATIONS TABLE
+-- ============================================================
+
+-- Add stage column to discovery locations (replaces status)
+DO $$
+BEGIN
+  IF NOT EXISTS (SELECT 1 FROM information_schema.columns
+    WHERE table_name = 'dutchie_discovery_locations' AND column_name = 'stage') THEN
+    ALTER TABLE dutchie_discovery_locations ADD COLUMN stage VARCHAR(20) DEFAULT 'discovered';
+  END IF;
+END $$;
+
+-- Migrate existing status values to stage
+UPDATE dutchie_discovery_locations
+SET stage = CASE
+  WHEN status = 'discovered' THEN 'discovered'
+  WHEN status = 'verified' THEN 'validated'
+  WHEN status = 'rejected' THEN 'failing'
+  WHEN status = 'merged' THEN 'validated'
+  ELSE 'discovered'
+END
+WHERE stage IS NULL OR stage = '';
+
+-- ============================================================
+-- UPDATE CRAWLER PROFILES TABLE
+-- ============================================================
+
+-- Ensure status column exists and update to new values
+UPDATE dispensary_crawler_profiles
+SET status = CASE
+  WHEN status = 'sandbox' THEN 'sandbox'
+  WHEN status = 'production' THEN 'production'
+  WHEN status = 'needs_manual' THEN 'failing'
+  WHEN status = 'failing' THEN 'failing'
+  WHEN status = 'disabled' THEN 'failing'
+  WHEN status IS NULL THEN 'promoted'
+  ELSE 'promoted'
+END;
+
+-- ============================================================
+-- ADD STAGE TRACKING TO DISPENSARIES
+-- ============================================================
+
+DO $$
+BEGIN
+  -- Add stage column to dispensaries for quick filtering
+  IF NOT EXISTS (SELECT 1 FROM information_schema.columns
+    WHERE table_name = 'dispensaries' AND column_name = 'stage') THEN
+    ALTER TABLE dispensaries ADD COLUMN stage VARCHAR(20) DEFAULT 'promoted';
+  END IF;
+
+  -- Add stage_changed_at for tracking
+  IF NOT EXISTS (SELECT 1 FROM information_schema.columns
+    WHERE table_name = 'dispensaries' AND column_name = 'stage_changed_at') THEN
+    ALTER TABLE dispensaries ADD COLUMN stage_changed_at TIMESTAMP WITH TIME ZONE DEFAULT CURRENT_TIMESTAMP;
+  END IF;
+
+  -- Add first_crawl_at to track sandbox → production transition
+  IF NOT EXISTS (SELECT 1 FROM information_schema.columns
+    WHERE table_name = 'dispensaries' AND column_name = 'first_crawl_at') THEN
+    ALTER TABLE dispensaries ADD COLUMN first_crawl_at TIMESTAMP WITH TIME ZONE;
+  END IF;
+
+  -- Add last_successful_crawl_at
+  IF NOT EXISTS (SELECT 1 FROM information_schema.columns
+    WHERE table_name = 'dispensaries' AND column_name = 'last_successful_crawl_at') THEN
+    ALTER TABLE dispensaries ADD COLUMN last_successful_crawl_at TIMESTAMP WITH TIME ZONE;
+  END IF;
+END $$;
+
+-- Set initial stage for existing dispensaries based on their crawler profile status
+UPDATE dispensaries d
+SET stage = COALESCE(
+  (SELECT dcp.status FROM dispensary_crawler_profiles dcp
+   WHERE dcp.dispensary_id = d.id AND dcp.enabled = true
+   ORDER BY dcp.updated_at DESC LIMIT 1),
+  'promoted'
+)
+WHERE d.stage IS NULL OR d.stage = '';
+
+-- ============================================================
+-- INDEXES FOR STAGE-BASED QUERIES
+-- ============================================================
+
+CREATE INDEX IF NOT EXISTS idx_dispensaries_stage ON dispensaries(stage);
+CREATE INDEX IF NOT EXISTS idx_dispensaries_stage_state ON dispensaries(stage, state);
+CREATE INDEX IF NOT EXISTS idx_discovery_locations_stage ON dutchie_discovery_locations(stage);
+CREATE INDEX IF NOT EXISTS idx_crawler_profiles_status ON dispensary_crawler_profiles(status);
+
+-- ============================================================
+-- STAGE TRANSITION LOG
+-- ============================================================
+
+CREATE TABLE IF NOT EXISTS stage_transitions (
+  id SERIAL PRIMARY KEY,
+
+  -- What changed
+  entity_type VARCHAR(20) NOT NULL, -- 'discovery_location' or 'dispensary'
+  entity_id INTEGER NOT NULL,
+
+  -- Stage change
+  from_stage VARCHAR(20),
+  to_stage VARCHAR(20) NOT NULL,
+
+  -- Context
+  trigger_type VARCHAR(50) NOT NULL, -- 'api', 'scheduler', 'manual', 'auto'
+  trigger_endpoint VARCHAR(200),
+
+  -- Outcome
+  success BOOLEAN DEFAULT TRUE,
+  error_message TEXT,
+  metadata JSONB,
+
+  -- Timing
+  duration_ms INTEGER,
+  created_at TIMESTAMP WITH TIME ZONE DEFAULT CURRENT_TIMESTAMP
+);
+
+CREATE INDEX IF NOT EXISTS idx_stage_transitions_entity ON stage_transitions(entity_type, entity_id);
+CREATE INDEX IF NOT EXISTS idx_stage_transitions_to_stage ON stage_transitions(to_stage);
+CREATE INDEX IF NOT EXISTS idx_stage_transitions_created ON stage_transitions(created_at DESC);
+
+-- ============================================================
+-- COMMENTS
+-- ============================================================
+
+COMMENT ON TABLE stage_transitions IS 'Audit log for all stage transitions in the pipeline';
+COMMENT ON COLUMN dispensaries.stage IS 'Current pipeline stage: discovered, validated, promoted, sandbox, production, failing';
+COMMENT ON COLUMN dispensaries.stage_changed_at IS 'When the stage was last changed';
+COMMENT ON COLUMN dispensaries.first_crawl_at IS 'When the first crawl was attempted (sandbox stage)';
+COMMENT ON COLUMN dispensaries.last_successful_crawl_at IS 'When the last successful crawl completed';

backend/migrations/070_product_variants.sql

@@ -0,0 +1,239 @@
+-- ============================================================================
+-- Migration 070: Product Variants Tables
+-- ============================================================================
+--
+-- Purpose: Store variant-level pricing and inventory as first-class entities
+--          to enable time-series analytics, price comparisons, and sale tracking.
+--
+-- Enables queries like:
+--   - Price history for a specific variant (1g Blue Dream over time)
+--   - Sale frequency analysis (how often is this on special?)
+--   - Cross-store price comparison (who has cheapest 1g flower?)
+--   - Current specials across all stores
+--
+-- RULES:
+--   - STRICTLY ADDITIVE (no DROP, DELETE, TRUNCATE)
+--   - All new tables use IF NOT EXISTS
+--   - All indexes use IF NOT EXISTS
+--
+-- ============================================================================
+
+-- ============================================================================
+-- SECTION 1: PRODUCT_VARIANTS TABLE (Current State)
+-- ============================================================================
+-- One row per product+option combination. Tracks current pricing/inventory.
+
+CREATE TABLE IF NOT EXISTS product_variants (
+    id SERIAL PRIMARY KEY,
+    store_product_id INTEGER NOT NULL REFERENCES store_products(id) ON DELETE CASCADE,
+    dispensary_id INTEGER NOT NULL REFERENCES dispensaries(id) ON DELETE CASCADE,
+
+    -- Variant identity (from Dutchie POSMetaData.children)
+    option VARCHAR(100) NOT NULL,              -- "1g", "3.5g", "1/8oz", "100mg"
+    canonical_sku VARCHAR(100),                -- Dutchie canonicalSKU
+    canonical_id VARCHAR(100),                 -- Dutchie canonicalID
+    canonical_name VARCHAR(500),               -- Dutchie canonicalName
+
+    -- Current pricing (in dollars, not cents)
+    price_rec NUMERIC(10,2),
+    price_med NUMERIC(10,2),
+    price_rec_special NUMERIC(10,2),
+    price_med_special NUMERIC(10,2),
+
+    -- Current inventory
+    quantity INTEGER,
+    quantity_available INTEGER,
+    in_stock BOOLEAN DEFAULT TRUE,
+
+    -- Special/sale status
+    is_on_special BOOLEAN DEFAULT FALSE,
+
+    -- Weight/size parsing (for analytics)
+    weight_value NUMERIC(10,2),               -- 1, 3.5, 28, etc.
+    weight_unit VARCHAR(20),                  -- g, oz, mg, ml, etc.
+
+    -- Timestamps
+    first_seen_at TIMESTAMPTZ DEFAULT NOW(),
+    last_seen_at TIMESTAMPTZ DEFAULT NOW(),
+    last_price_change_at TIMESTAMPTZ,
+    last_stock_change_at TIMESTAMPTZ,
+
+    created_at TIMESTAMPTZ DEFAULT NOW(),
+    updated_at TIMESTAMPTZ DEFAULT NOW(),
+
+    UNIQUE(store_product_id, option)
+);
+
+-- Indexes for common queries
+CREATE INDEX IF NOT EXISTS idx_variants_store_product ON product_variants(store_product_id);
+CREATE INDEX IF NOT EXISTS idx_variants_dispensary ON product_variants(dispensary_id);
+CREATE INDEX IF NOT EXISTS idx_variants_option ON product_variants(option);
+CREATE INDEX IF NOT EXISTS idx_variants_in_stock ON product_variants(dispensary_id, in_stock) WHERE in_stock = TRUE;
+CREATE INDEX IF NOT EXISTS idx_variants_on_special ON product_variants(dispensary_id, is_on_special) WHERE is_on_special = TRUE;
+CREATE INDEX IF NOT EXISTS idx_variants_canonical_sku ON product_variants(canonical_sku) WHERE canonical_sku IS NOT NULL;
+CREATE INDEX IF NOT EXISTS idx_variants_price_rec ON product_variants(price_rec) WHERE price_rec IS NOT NULL;
+
+COMMENT ON TABLE product_variants IS 'Current state of each product variant (weight/size option). One row per product+option.';
+COMMENT ON COLUMN product_variants.option IS 'Weight/size option string from Dutchie (e.g., "1g", "3.5g", "1/8oz")';
+COMMENT ON COLUMN product_variants.canonical_sku IS 'Dutchie POS SKU for cross-store matching';
+
+
+-- ============================================================================
+-- SECTION 2: PRODUCT_VARIANT_SNAPSHOTS TABLE (Historical Data)
+-- ============================================================================
+-- Time-series data for variant pricing. One row per variant per crawl.
+-- CRITICAL: NEVER DELETE from this table.
+
+CREATE TABLE IF NOT EXISTS product_variant_snapshots (
+    id SERIAL PRIMARY KEY,
+    product_variant_id INTEGER NOT NULL REFERENCES product_variants(id) ON DELETE CASCADE,
+    store_product_id INTEGER REFERENCES store_products(id) ON DELETE SET NULL,
+    dispensary_id INTEGER NOT NULL REFERENCES dispensaries(id) ON DELETE CASCADE,
+    crawl_run_id INTEGER REFERENCES crawl_runs(id) ON DELETE SET NULL,
+
+    -- Variant identity (denormalized for query performance)
+    option VARCHAR(100) NOT NULL,
+
+    -- Pricing at time of capture
+    price_rec NUMERIC(10,2),
+    price_med NUMERIC(10,2),
+    price_rec_special NUMERIC(10,2),
+    price_med_special NUMERIC(10,2),
+
+    -- Inventory at time of capture
+    quantity INTEGER,
+    in_stock BOOLEAN DEFAULT TRUE,
+
+    -- Special status at time of capture
+    is_on_special BOOLEAN DEFAULT FALSE,
+
+    -- Feed presence (FALSE = variant missing from crawl)
+    is_present_in_feed BOOLEAN DEFAULT TRUE,
+
+    -- Capture timestamp
+    captured_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
+
+    created_at TIMESTAMPTZ DEFAULT NOW()
+);
+
+-- Indexes for time-series queries
+CREATE INDEX IF NOT EXISTS idx_variant_snapshots_variant ON product_variant_snapshots(product_variant_id, captured_at DESC);
+CREATE INDEX IF NOT EXISTS idx_variant_snapshots_dispensary ON product_variant_snapshots(dispensary_id, captured_at DESC);
+CREATE INDEX IF NOT EXISTS idx_variant_snapshots_crawl ON product_variant_snapshots(crawl_run_id) WHERE crawl_run_id IS NOT NULL;
+CREATE INDEX IF NOT EXISTS idx_variant_snapshots_captured ON product_variant_snapshots(captured_at DESC);
+CREATE INDEX IF NOT EXISTS idx_variant_snapshots_special ON product_variant_snapshots(is_on_special, captured_at DESC) WHERE is_on_special = TRUE;
+CREATE INDEX IF NOT EXISTS idx_variant_snapshots_option ON product_variant_snapshots(option, captured_at DESC);
+
+COMMENT ON TABLE product_variant_snapshots IS 'Historical variant pricing/inventory. One row per variant per crawl. NEVER DELETE.';
+
+
+-- ============================================================================
+-- SECTION 3: USEFUL VIEWS
+-- ============================================================================
+
+-- View: Current specials across all stores
+CREATE OR REPLACE VIEW v_current_specials AS
+SELECT
+    pv.id as variant_id,
+    sp.id as product_id,
+    sp.name_raw as product_name,
+    sp.brand_name_raw as brand_name,
+    sp.category_raw as category,
+    d.id as dispensary_id,
+    d.name as dispensary_name,
+    d.city,
+    d.state,
+    pv.option,
+    pv.price_rec,
+    pv.price_rec_special,
+    ROUND(((pv.price_rec - pv.price_rec_special) / NULLIF(pv.price_rec, 0)) * 100, 1) as discount_percent,
+    pv.quantity,
+    pv.in_stock,
+    pv.last_seen_at
+FROM product_variants pv
+JOIN store_products sp ON sp.id = pv.store_product_id
+JOIN dispensaries d ON d.id = pv.dispensary_id
+WHERE pv.is_on_special = TRUE
+  AND pv.in_stock = TRUE
+  AND pv.price_rec_special IS NOT NULL
+  AND pv.price_rec_special < pv.price_rec;
+
+COMMENT ON VIEW v_current_specials IS 'All products currently on special across all stores';
+
+
+-- View: Price comparison for a product across stores
+CREATE OR REPLACE VIEW v_price_comparison AS
+SELECT
+    sp.name_raw as product_name,
+    sp.brand_name_raw as brand_name,
+    sp.category_raw as category,
+    pv.option,
+    d.id as dispensary_id,
+    d.name as dispensary_name,
+    d.city,
+    pv.price_rec,
+    pv.price_rec_special,
+    pv.is_on_special,
+    pv.in_stock,
+    pv.quantity,
+    RANK() OVER (PARTITION BY sp.name_raw, pv.option ORDER BY COALESCE(pv.price_rec_special, pv.price_rec) ASC) as price_rank
+FROM product_variants pv
+JOIN store_products sp ON sp.id = pv.store_product_id
+JOIN dispensaries d ON d.id = pv.dispensary_id
+WHERE pv.in_stock = TRUE
+  AND (pv.price_rec IS NOT NULL OR pv.price_rec_special IS NOT NULL);
+
+COMMENT ON VIEW v_price_comparison IS 'Compare prices for same product across stores, ranked by price';
+
+
+-- View: Latest snapshot per variant
+CREATE OR REPLACE VIEW v_latest_variant_snapshots AS
+SELECT DISTINCT ON (product_variant_id)
+    pvs.*
+FROM product_variant_snapshots pvs
+ORDER BY product_variant_id, captured_at DESC;
+
+
+-- ============================================================================
+-- SECTION 4: HELPER FUNCTION FOR SALE FREQUENCY
+-- ============================================================================
+
+-- Function to calculate sale frequency for a variant
+CREATE OR REPLACE FUNCTION get_variant_sale_stats(p_variant_id INTEGER, p_days INTEGER DEFAULT 30)
+RETURNS TABLE (
+    total_snapshots BIGINT,
+    times_on_special BIGINT,
+    special_frequency_pct NUMERIC,
+    avg_discount_pct NUMERIC,
+    min_price NUMERIC,
+    max_price NUMERIC,
+    avg_price NUMERIC
+) AS $$
+BEGIN
+    RETURN QUERY
+    SELECT
+        COUNT(*)::BIGINT as total_snapshots,
+        COUNT(*) FILTER (WHERE is_on_special)::BIGINT as times_on_special,
+        ROUND((COUNT(*) FILTER (WHERE is_on_special)::NUMERIC / NULLIF(COUNT(*), 0)) * 100, 1) as special_frequency_pct,
+        ROUND(AVG(
+            CASE WHEN is_on_special AND price_rec_special IS NOT NULL AND price_rec IS NOT NULL
+            THEN ((price_rec - price_rec_special) / NULLIF(price_rec, 0)) * 100
+            END
+        ), 1) as avg_discount_pct,
+        MIN(COALESCE(price_rec_special, price_rec)) as min_price,
+        MAX(price_rec) as max_price,
+        ROUND(AVG(COALESCE(price_rec_special, price_rec)), 2) as avg_price
+    FROM product_variant_snapshots
+    WHERE product_variant_id = p_variant_id
+      AND captured_at >= NOW() - (p_days || ' days')::INTERVAL;
+END;
+$$ LANGUAGE plpgsql;
+
+COMMENT ON FUNCTION get_variant_sale_stats IS 'Get sale frequency and price stats for a variant over N days';
+
+
+-- ============================================================================
+-- DONE
+-- ============================================================================
+
+SELECT 'Migration 070 completed. Product variants tables ready for time-series analytics.' AS status;

backend/migrations/071_harmonize_store_products.sql

@@ -0,0 +1,53 @@
+-- Migration 071: Harmonize store_products with dutchie_products
+-- Adds missing columns to store_products to consolidate on a single canonical table
+
+-- Product details
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS description TEXT;
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS weight VARCHAR(50);
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS weights JSONB;
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS measurements JSONB;
+
+-- Cannabinoid/terpene data
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS effects JSONB;
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS terpenes JSONB;
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS cannabinoids_v2 JSONB;
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS thc_content NUMERIC(10,4);
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS cbd_content NUMERIC(10,4);
+
+-- Images
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS images JSONB;
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS primary_image_url TEXT;
+
+-- Inventory
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS total_quantity_available INTEGER DEFAULT 0;
+
+-- Status/flags
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS status VARCHAR(50);
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS featured BOOLEAN DEFAULT FALSE;
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS coming_soon BOOLEAN DEFAULT FALSE;
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS visibility_lost BOOLEAN DEFAULT FALSE;
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS visibility_lost_at TIMESTAMP WITH TIME ZONE;
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS visibility_restored_at TIMESTAMP WITH TIME ZONE;
+
+-- Threshold flags (Dutchie-specific)
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS is_below_threshold BOOLEAN DEFAULT FALSE;
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS is_below_kiosk_threshold BOOLEAN DEFAULT FALSE;
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS options_below_threshold BOOLEAN DEFAULT FALSE;
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS options_below_kiosk_threshold BOOLEAN DEFAULT FALSE;
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS certificate_of_analysis_enabled BOOLEAN DEFAULT FALSE;
+
+-- Platform metadata
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS external_product_id VARCHAR(100);
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS c_name VARCHAR(500);
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS past_c_names TEXT[];
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS latest_raw_payload JSONB;
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS created_at_platform TIMESTAMP WITH TIME ZONE;
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS updated_at_platform TIMESTAMP WITH TIME ZONE;
+
+-- Indexes for common queries
+CREATE INDEX IF NOT EXISTS idx_store_products_external_id ON store_products(external_product_id);
+CREATE INDEX IF NOT EXISTS idx_store_products_visibility_lost ON store_products(visibility_lost) WHERE visibility_lost = TRUE;
+CREATE INDEX IF NOT EXISTS idx_store_products_status ON store_products(status);
+
+-- Add comment
+COMMENT ON TABLE store_products IS 'Canonical product table - consolidated from dutchie_products';

backend/migrations/072_product_views.sql

@@ -0,0 +1,74 @@
+-- Migration 072: Create compatibility views for store_products and store_product_snapshots
+-- These views provide backward-compatible column names for API routes
+
+-- v_products view - aliases store_products columns to match legacy dutchie_products naming
+CREATE OR REPLACE VIEW v_products AS
+SELECT
+  id,
+  dispensary_id,
+  provider_product_id as external_product_id,
+  provider_product_id as dutchie_id,
+  name_raw as name,
+  brand_name_raw as brand_name,
+  category_raw as type,
+  subcategory_raw as subcategory,
+  strain_type,
+  thc_percent as thc,
+  cbd_percent as cbd,
+  stock_status,
+  is_in_stock,
+  stock_quantity,
+  image_url,
+  primary_image_url,
+  images,
+  effects,
+  description,
+  is_on_special,
+  featured,
+  medical_only,
+  rec_only,
+  external_product_id as external_id,
+  provider,
+  created_at,
+  updated_at
+FROM store_products;
+
+-- v_product_snapshots view - aliases store_product_snapshots columns to match legacy naming
+CREATE OR REPLACE VIEW v_product_snapshots AS
+SELECT
+  id,
+  store_product_id,
+  dispensary_id,
+  provider,
+  provider_product_id,
+  crawl_run_id,
+  captured_at as crawled_at,
+  name_raw,
+  brand_name_raw,
+  category_raw,
+  subcategory_raw,
+  -- Convert price_rec (dollars) to rec_min_price_cents (cents)
+  CASE WHEN price_rec IS NOT NULL THEN (price_rec * 100)::integer END as rec_min_price_cents,
+  CASE WHEN price_rec IS NOT NULL THEN (price_rec * 100)::integer END as rec_max_price_cents,
+  CASE WHEN price_rec_special IS NOT NULL THEN (price_rec_special * 100)::integer END as rec_min_special_price_cents,
+  CASE WHEN price_med IS NOT NULL THEN (price_med * 100)::integer END as med_min_price_cents,
+  CASE WHEN price_med IS NOT NULL THEN (price_med * 100)::integer END as med_max_price_cents,
+  CASE WHEN price_med_special IS NOT NULL THEN (price_med_special * 100)::integer END as med_min_special_price_cents,
+  is_on_special as special,
+  discount_percent,
+  is_in_stock,
+  stock_quantity,
+  stock_status,
+  stock_quantity as total_quantity_available,
+  thc_percent,
+  cbd_percent,
+  image_url,
+  raw_data as options,
+  created_at
+FROM store_product_snapshots;
+
+-- Add indexes for the views' underlying tables
+CREATE INDEX IF NOT EXISTS idx_store_products_dispensary ON store_products(dispensary_id);
+CREATE INDEX IF NOT EXISTS idx_store_products_stock ON store_products(stock_status);
+CREATE INDEX IF NOT EXISTS idx_store_snapshots_product ON store_product_snapshots(store_product_id);
+CREATE INDEX IF NOT EXISTS idx_store_snapshots_captured ON store_product_snapshots(captured_at DESC);

backend/migrations/073_proxy_timezone.sql

@@ -0,0 +1,12 @@
+-- Add timezone column to proxies table for geo-consistent fingerprinting
+-- This allows matching Accept-Language and other headers to proxy location
+
+ALTER TABLE proxies
+ADD COLUMN IF NOT EXISTS timezone VARCHAR(50);
+
+-- Add timezone to failed_proxies as well
+ALTER TABLE failed_proxies
+ADD COLUMN IF NOT EXISTS timezone VARCHAR(50);
+
+-- Comment explaining usage
+COMMENT ON COLUMN proxies.timezone IS 'IANA timezone (e.g., America/Phoenix) for geo-consistent fingerprinting';

backend/migrations/074_worker_commands.sql

@@ -0,0 +1,27 @@
+-- Migration: Worker Commands Table
+-- Purpose: Store commands for workers (decommission, etc.)
+-- Workers poll this table after each task to check for commands
+
+CREATE TABLE IF NOT EXISTS worker_commands (
+  id SERIAL PRIMARY KEY,
+  worker_id TEXT NOT NULL,
+  command TEXT NOT NULL,  -- 'decommission', 'pause', 'resume'
+  reason TEXT,
+  issued_by TEXT,
+  issued_at TIMESTAMPTZ DEFAULT NOW(),
+  acknowledged_at TIMESTAMPTZ,
+  executed_at TIMESTAMPTZ,
+  status TEXT DEFAULT 'pending'  -- 'pending', 'acknowledged', 'executed', 'cancelled'
+);
+
+-- Index for worker lookups
+CREATE INDEX IF NOT EXISTS idx_worker_commands_worker_id ON worker_commands(worker_id);
+CREATE INDEX IF NOT EXISTS idx_worker_commands_pending ON worker_commands(worker_id, status) WHERE status = 'pending';
+
+-- Add decommission_requested column to worker_registry for quick checks
+ALTER TABLE worker_registry ADD COLUMN IF NOT EXISTS decommission_requested BOOLEAN DEFAULT FALSE;
+ALTER TABLE worker_registry ADD COLUMN IF NOT EXISTS decommission_reason TEXT;
+ALTER TABLE worker_registry ADD COLUMN IF NOT EXISTS decommission_requested_at TIMESTAMPTZ;
+
+-- Comment
+COMMENT ON TABLE worker_commands IS 'Commands issued to workers (decommission after task, pause, etc.)';

backend/migrations/074_worker_task_queue.sql

@@ -0,0 +1,322 @@
+-- Migration 074: Worker Task Queue System
+-- Implements role-based task queue with per-store locking and capacity tracking
+
+-- Task queue table
+CREATE TABLE IF NOT EXISTS worker_tasks (
+  id SERIAL PRIMARY KEY,
+
+  -- Task identification
+  role VARCHAR(50) NOT NULL,           -- store_discovery, entry_point_discovery, product_discovery, product_resync, analytics_refresh
+  dispensary_id INTEGER REFERENCES dispensaries(id) ON DELETE CASCADE,
+  platform VARCHAR(20),                 -- dutchie, jane, treez, etc.
+
+  -- Task state
+  status VARCHAR(20) NOT NULL DEFAULT 'pending',
+  priority INTEGER DEFAULT 0,           -- Higher = more urgent
+
+  -- Scheduling
+  scheduled_for TIMESTAMPTZ,            -- For batch scheduling (e.g., every 4 hours)
+
+  -- Ownership
+  worker_id VARCHAR(100),               -- Pod name or worker ID
+  claimed_at TIMESTAMPTZ,
+  started_at TIMESTAMPTZ,
+  completed_at TIMESTAMPTZ,
+  last_heartbeat_at TIMESTAMPTZ,
+
+  -- Results
+  result JSONB,                         -- Task output data
+  error_message TEXT,
+  retry_count INTEGER DEFAULT 0,
+  max_retries INTEGER DEFAULT 3,
+
+  -- Metadata
+  created_at TIMESTAMPTZ DEFAULT NOW(),
+  updated_at TIMESTAMPTZ DEFAULT NOW(),
+
+  -- Constraints
+  CONSTRAINT valid_status CHECK (status IN ('pending', 'claimed', 'running', 'completed', 'failed', 'stale'))
+);
+
+-- Indexes for efficient task claiming
+CREATE INDEX IF NOT EXISTS idx_worker_tasks_pending
+  ON worker_tasks(role, priority DESC, created_at ASC)
+  WHERE status = 'pending';
+
+CREATE INDEX IF NOT EXISTS idx_worker_tasks_claimed
+  ON worker_tasks(worker_id, claimed_at)
+  WHERE status = 'claimed';
+
+CREATE INDEX IF NOT EXISTS idx_worker_tasks_running
+  ON worker_tasks(worker_id, last_heartbeat_at)
+  WHERE status = 'running';
+
+CREATE INDEX IF NOT EXISTS idx_worker_tasks_dispensary
+  ON worker_tasks(dispensary_id)
+  WHERE dispensary_id IS NOT NULL;
+
+CREATE INDEX IF NOT EXISTS idx_worker_tasks_scheduled
+  ON worker_tasks(scheduled_for)
+  WHERE status = 'pending' AND scheduled_for IS NOT NULL;
+
+CREATE INDEX IF NOT EXISTS idx_worker_tasks_history
+  ON worker_tasks(role, completed_at DESC)
+  WHERE status IN ('completed', 'failed');
+
+-- Partial unique index to prevent duplicate active tasks per store
+-- Only one task can be claimed/running for a given dispensary at a time
+CREATE UNIQUE INDEX IF NOT EXISTS idx_worker_tasks_unique_active_store
+  ON worker_tasks(dispensary_id)
+  WHERE status IN ('claimed', 'running') AND dispensary_id IS NOT NULL;
+
+-- Worker registration table (tracks active workers)
+CREATE TABLE IF NOT EXISTS worker_registry (
+  id SERIAL PRIMARY KEY,
+  worker_id VARCHAR(100) UNIQUE NOT NULL,
+  role VARCHAR(50) NOT NULL,
+  pod_name VARCHAR(100),
+  hostname VARCHAR(100),
+  started_at TIMESTAMPTZ DEFAULT NOW(),
+  last_heartbeat_at TIMESTAMPTZ DEFAULT NOW(),
+  tasks_completed INTEGER DEFAULT 0,
+  tasks_failed INTEGER DEFAULT 0,
+  status VARCHAR(20) DEFAULT 'active',
+
+  CONSTRAINT valid_worker_status CHECK (status IN ('active', 'idle', 'offline'))
+);
+
+CREATE INDEX IF NOT EXISTS idx_worker_registry_role
+  ON worker_registry(role, status);
+
+CREATE INDEX IF NOT EXISTS idx_worker_registry_heartbeat
+  ON worker_registry(last_heartbeat_at)
+  WHERE status = 'active';
+
+-- Task completion tracking (summarized history)
+CREATE TABLE IF NOT EXISTS task_completion_log (
+  id SERIAL PRIMARY KEY,
+  role VARCHAR(50) NOT NULL,
+  date DATE NOT NULL DEFAULT CURRENT_DATE,
+  hour INTEGER NOT NULL DEFAULT EXTRACT(HOUR FROM NOW()),
+
+  tasks_created INTEGER DEFAULT 0,
+  tasks_completed INTEGER DEFAULT 0,
+  tasks_failed INTEGER DEFAULT 0,
+
+  avg_duration_sec NUMERIC(10,2),
+  min_duration_sec NUMERIC(10,2),
+  max_duration_sec NUMERIC(10,2),
+
+  updated_at TIMESTAMPTZ DEFAULT NOW(),
+
+  UNIQUE(role, date, hour)
+);
+
+-- Capacity planning view
+CREATE OR REPLACE VIEW v_worker_capacity AS
+SELECT
+  role,
+  COUNT(*) FILTER (WHERE status = 'pending') as pending_tasks,
+  COUNT(*) FILTER (WHERE status = 'pending' AND (scheduled_for IS NULL OR scheduled_for <= NOW())) as ready_tasks,
+  COUNT(*) FILTER (WHERE status = 'claimed') as claimed_tasks,
+  COUNT(*) FILTER (WHERE status = 'running') as running_tasks,
+  COUNT(*) FILTER (WHERE status = 'completed' AND completed_at > NOW() - INTERVAL '1 hour') as completed_last_hour,
+  COUNT(*) FILTER (WHERE status = 'failed' AND completed_at > NOW() - INTERVAL '1 hour') as failed_last_hour,
+  COUNT(DISTINCT worker_id) FILTER (WHERE status IN ('claimed', 'running')) as active_workers,
+  AVG(EXTRACT(EPOCH FROM (completed_at - started_at)))
+    FILTER (WHERE status = 'completed' AND completed_at > NOW() - INTERVAL '1 hour') as avg_duration_sec,
+  -- Capacity planning metrics
+  CASE
+    WHEN COUNT(*) FILTER (WHERE status = 'completed' AND completed_at > NOW() - INTERVAL '1 hour') > 0
+    THEN 3600.0 / NULLIF(AVG(EXTRACT(EPOCH FROM (completed_at - started_at)))
+      FILTER (WHERE status = 'completed' AND completed_at > NOW() - INTERVAL '1 hour'), 0)
+    ELSE NULL
+  END as tasks_per_worker_hour,
+  -- Estimated time to drain queue
+  CASE
+    WHEN COUNT(DISTINCT worker_id) FILTER (WHERE status IN ('claimed', 'running')) > 0
+      AND COUNT(*) FILTER (WHERE status = 'completed' AND completed_at > NOW() - INTERVAL '1 hour') > 0
+    THEN COUNT(*) FILTER (WHERE status = 'pending') / NULLIF(
+      COUNT(DISTINCT worker_id) FILTER (WHERE status IN ('claimed', 'running')) *
+      (3600.0 / NULLIF(AVG(EXTRACT(EPOCH FROM (completed_at - started_at)))
+        FILTER (WHERE status = 'completed' AND completed_at > NOW() - INTERVAL '1 hour'), 0)),
+      0
+    )
+    ELSE NULL
+  END as estimated_hours_to_drain
+FROM worker_tasks
+GROUP BY role;
+
+-- Task history view (for UI)
+CREATE OR REPLACE VIEW v_task_history AS
+SELECT
+  t.id,
+  t.role,
+  t.dispensary_id,
+  d.name as dispensary_name,
+  t.platform,
+  t.status,
+  t.priority,
+  t.worker_id,
+  t.scheduled_for,
+  t.claimed_at,
+  t.started_at,
+  t.completed_at,
+  t.error_message,
+  t.retry_count,
+  t.created_at,
+  EXTRACT(EPOCH FROM (t.completed_at - t.started_at)) as duration_sec
+FROM worker_tasks t
+LEFT JOIN dispensaries d ON d.id = t.dispensary_id
+ORDER BY t.created_at DESC;
+
+-- Function to claim a task atomically
+CREATE OR REPLACE FUNCTION claim_task(
+  p_role VARCHAR(50),
+  p_worker_id VARCHAR(100)
+) RETURNS worker_tasks AS $$
+DECLARE
+  claimed_task worker_tasks;
+BEGIN
+  UPDATE worker_tasks
+  SET
+    status = 'claimed',
+    worker_id = p_worker_id,
+    claimed_at = NOW(),
+    updated_at = NOW()
+  WHERE id = (
+    SELECT id FROM worker_tasks
+    WHERE role = p_role
+      AND status = 'pending'
+      AND (scheduled_for IS NULL OR scheduled_for <= NOW())
+      -- Exclude stores that already have an active task
+      AND (dispensary_id IS NULL OR dispensary_id NOT IN (
+        SELECT dispensary_id FROM worker_tasks
+        WHERE status IN ('claimed', 'running')
+        AND dispensary_id IS NOT NULL
+      ))
+    ORDER BY priority DESC, created_at ASC
+    LIMIT 1
+    FOR UPDATE SKIP LOCKED
+  )
+  RETURNING * INTO claimed_task;
+
+  RETURN claimed_task;
+END;
+$$ LANGUAGE plpgsql;
+
+-- Function to mark stale tasks (workers that died)
+CREATE OR REPLACE FUNCTION recover_stale_tasks(
+  stale_threshold_minutes INTEGER DEFAULT 10
+) RETURNS INTEGER AS $$
+DECLARE
+  recovered_count INTEGER;
+BEGIN
+  WITH stale AS (
+    UPDATE worker_tasks
+    SET
+      status = 'pending',
+      worker_id = NULL,
+      claimed_at = NULL,
+      started_at = NULL,
+      retry_count = retry_count + 1,
+      updated_at = NOW()
+    WHERE status IN ('claimed', 'running')
+      AND last_heartbeat_at < NOW() - (stale_threshold_minutes || ' minutes')::INTERVAL
+      AND retry_count < max_retries
+    RETURNING id
+  )
+  SELECT COUNT(*) INTO recovered_count FROM stale;
+
+  -- Mark tasks that exceeded retries as failed
+  UPDATE worker_tasks
+  SET
+    status = 'failed',
+    error_message = 'Exceeded max retries after worker failures',
+    completed_at = NOW(),
+    updated_at = NOW()
+  WHERE status IN ('claimed', 'running')
+    AND last_heartbeat_at < NOW() - (stale_threshold_minutes || ' minutes')::INTERVAL
+    AND retry_count >= max_retries;
+
+  RETURN recovered_count;
+END;
+$$ LANGUAGE plpgsql;
+
+-- Function to generate daily resync tasks
+CREATE OR REPLACE FUNCTION generate_resync_tasks(
+  p_batches_per_day INTEGER DEFAULT 6,  -- Every 4 hours
+  p_date DATE DEFAULT CURRENT_DATE
+) RETURNS INTEGER AS $$
+DECLARE
+  store_count INTEGER;
+  stores_per_batch INTEGER;
+  batch_num INTEGER;
+  scheduled_time TIMESTAMPTZ;
+  created_count INTEGER := 0;
+BEGIN
+  -- Count active stores that need resync
+  SELECT COUNT(*) INTO store_count
+  FROM dispensaries
+  WHERE crawl_enabled = true
+    AND menu_type = 'dutchie'
+    AND platform_dispensary_id IS NOT NULL;
+
+  IF store_count = 0 THEN
+    RETURN 0;
+  END IF;
+
+  stores_per_batch := CEIL(store_count::NUMERIC / p_batches_per_day);
+
+  FOR batch_num IN 0..(p_batches_per_day - 1) LOOP
+    scheduled_time := p_date + (batch_num * 4 || ' hours')::INTERVAL;
+
+    INSERT INTO worker_tasks (role, dispensary_id, platform, scheduled_for, priority)
+    SELECT
+      'product_resync',
+      d.id,
+      'dutchie',
+      scheduled_time,
+      0
+    FROM (
+      SELECT id, ROW_NUMBER() OVER (ORDER BY id) as rn
+      FROM dispensaries
+      WHERE crawl_enabled = true
+        AND menu_type = 'dutchie'
+        AND platform_dispensary_id IS NOT NULL
+    ) d
+    WHERE d.rn > (batch_num * stores_per_batch)
+      AND d.rn <= ((batch_num + 1) * stores_per_batch)
+    ON CONFLICT DO NOTHING;
+
+    GET DIAGNOSTICS created_count = created_count + ROW_COUNT;
+  END LOOP;
+
+  RETURN created_count;
+END;
+$$ LANGUAGE plpgsql;
+
+-- Trigger to update timestamp
+CREATE OR REPLACE FUNCTION update_worker_tasks_timestamp()
+RETURNS TRIGGER AS $$
+BEGIN
+  NEW.updated_at = NOW();
+  RETURN NEW;
+END;
+$$ LANGUAGE plpgsql;
+
+DROP TRIGGER IF EXISTS worker_tasks_updated_at ON worker_tasks;
+CREATE TRIGGER worker_tasks_updated_at
+  BEFORE UPDATE ON worker_tasks
+  FOR EACH ROW
+  EXECUTE FUNCTION update_worker_tasks_timestamp();
+
+-- Comments
+COMMENT ON TABLE worker_tasks IS 'Central task queue for all worker roles';
+COMMENT ON TABLE worker_registry IS 'Registry of active workers and their stats';
+COMMENT ON TABLE task_completion_log IS 'Hourly aggregated task completion metrics';
+COMMENT ON VIEW v_worker_capacity IS 'Real-time capacity planning metrics per role';
+COMMENT ON VIEW v_task_history IS 'Task history with dispensary details for UI';
+COMMENT ON FUNCTION claim_task IS 'Atomically claim a task for a worker, respecting per-store locking';
+COMMENT ON FUNCTION recover_stale_tasks IS 'Release tasks from dead workers back to pending';
+COMMENT ON FUNCTION generate_resync_tasks IS 'Generate daily product resync tasks in batches';

backend/migrations/075_consecutive_misses.sql

@@ -0,0 +1,13 @@
+-- Migration 075: Add consecutive_misses column to store_products
+-- Used to track how many consecutive crawls a product has been missing from the feed
+-- After 3 consecutive misses, product is marked as OOS
+
+ALTER TABLE store_products
+ADD COLUMN IF NOT EXISTS consecutive_misses INTEGER NOT NULL DEFAULT 0;
+
+-- Index for finding products that need OOS check
+CREATE INDEX IF NOT EXISTS idx_store_products_consecutive_misses
+ON store_products (dispensary_id, consecutive_misses)
+WHERE consecutive_misses > 0;
+
+COMMENT ON COLUMN store_products.consecutive_misses IS 'Number of consecutive crawls where product was not in feed. Reset to 0 when seen. At 3, mark OOS.';

backend/migrations/076_visitor_analytics.sql

@@ -0,0 +1,71 @@
+-- Visitor location analytics for Findagram
+-- Tracks visitor locations to understand popular areas
+
+CREATE TABLE IF NOT EXISTS visitor_locations (
+  id SERIAL PRIMARY KEY,
+
+  -- Location data (from IP lookup)
+  ip_hash VARCHAR(64),           -- Hashed IP for privacy (SHA256)
+  city VARCHAR(100),
+  state VARCHAR(100),
+  state_code VARCHAR(10),
+  country VARCHAR(100),
+  country_code VARCHAR(10),
+  latitude DECIMAL(10, 7),
+  longitude DECIMAL(10, 7),
+
+  -- Visit metadata
+  domain VARCHAR(50) NOT NULL,    -- 'findagram.co', 'findadispo.com', etc.
+  page_path VARCHAR(255),         -- '/products', '/dispensaries/123', etc.
+  referrer VARCHAR(500),
+  user_agent VARCHAR(500),
+
+  -- Session tracking
+  session_id VARCHAR(64),         -- For grouping page views in a session
+
+  -- Timestamps
+  created_at TIMESTAMPTZ DEFAULT NOW()
+);
+
+-- Indexes for analytics queries
+CREATE INDEX IF NOT EXISTS idx_visitor_locations_domain ON visitor_locations(domain);
+CREATE INDEX IF NOT EXISTS idx_visitor_locations_city_state ON visitor_locations(city, state_code);
+CREATE INDEX IF NOT EXISTS idx_visitor_locations_created_at ON visitor_locations(created_at);
+CREATE INDEX IF NOT EXISTS idx_visitor_locations_session ON visitor_locations(session_id);
+
+-- Aggregated daily stats (materialized for performance)
+CREATE TABLE IF NOT EXISTS visitor_location_stats (
+  id SERIAL PRIMARY KEY,
+  date DATE NOT NULL,
+  domain VARCHAR(50) NOT NULL,
+  city VARCHAR(100),
+  state VARCHAR(100),
+  state_code VARCHAR(10),
+  country_code VARCHAR(10),
+
+  -- Metrics
+  visit_count INTEGER DEFAULT 0,
+  unique_sessions INTEGER DEFAULT 0,
+
+  UNIQUE(date, domain, city, state_code, country_code)
+);
+
+CREATE INDEX IF NOT EXISTS idx_visitor_stats_date ON visitor_location_stats(date);
+CREATE INDEX IF NOT EXISTS idx_visitor_stats_domain ON visitor_location_stats(domain);
+CREATE INDEX IF NOT EXISTS idx_visitor_stats_state ON visitor_location_stats(state_code);
+
+-- View for easy querying of top locations
+CREATE OR REPLACE VIEW v_top_visitor_locations AS
+SELECT
+  domain,
+  city,
+  state,
+  state_code,
+  country_code,
+  COUNT(*) as total_visits,
+  COUNT(DISTINCT session_id) as unique_sessions,
+  MAX(created_at) as last_visit
+FROM visitor_locations
+WHERE created_at > NOW() - INTERVAL '30 days'
+GROUP BY domain, city, state, state_code, country_code
+ORDER BY total_visits DESC;

backend/migrations/076_worker_registry.sql

@@ -0,0 +1,141 @@
+-- Migration 076: Worker Registry for Dynamic Workers
+-- Workers register on startup, receive a friendly name, and report heartbeats
+
+-- Name pool for workers (expandable, no hardcoding)
+CREATE TABLE IF NOT EXISTS worker_name_pool (
+  id SERIAL PRIMARY KEY,
+  name VARCHAR(50) UNIQUE NOT NULL,
+  in_use BOOLEAN DEFAULT FALSE,
+  assigned_to VARCHAR(100), -- worker_id
+  assigned_at TIMESTAMPTZ,
+  created_at TIMESTAMPTZ DEFAULT NOW()
+);
+
+-- Seed with initial names (can add more via API)
+INSERT INTO worker_name_pool (name) VALUES
+  ('Alice'), ('Bella'), ('Clara'), ('Diana'), ('Elena'),
+  ('Fiona'), ('Grace'), ('Hazel'), ('Iris'), ('Julia'),
+  ('Katie'), ('Luna'), ('Mia'), ('Nora'), ('Olive'),
+  ('Pearl'), ('Quinn'), ('Rosa'), ('Sara'), ('Tara'),
+  ('Uma'), ('Vera'), ('Wendy'), ('Xena'), ('Yuki'), ('Zara'),
+  ('Amber'), ('Blake'), ('Coral'), ('Dawn'), ('Echo'),
+  ('Fleur'), ('Gem'), ('Haven'), ('Ivy'), ('Jade'),
+  ('Kira'), ('Lotus'), ('Maple'), ('Nova'), ('Onyx'),
+  ('Pixel'), ('Quest'), ('Raven'), ('Sage'), ('Terra'),
+  ('Unity'), ('Violet'), ('Willow'), ('Xylo'), ('Yara'), ('Zen')
+ON CONFLICT (name) DO NOTHING;
+
+-- Worker registry - tracks active workers
+CREATE TABLE IF NOT EXISTS worker_registry (
+  id SERIAL PRIMARY KEY,
+  worker_id VARCHAR(100) UNIQUE NOT NULL,  -- e.g., "pod-abc123" or uuid
+  friendly_name VARCHAR(50),                -- assigned from pool
+  role VARCHAR(50) NOT NULL,                -- task role
+  pod_name VARCHAR(100),                    -- k8s pod name
+  hostname VARCHAR(100),                    -- machine hostname
+  ip_address VARCHAR(50),                   -- worker IP
+  status VARCHAR(20) DEFAULT 'starting',    -- starting, active, idle, offline, terminated
+  started_at TIMESTAMPTZ DEFAULT NOW(),
+  last_heartbeat_at TIMESTAMPTZ DEFAULT NOW(),
+  last_task_at TIMESTAMPTZ,
+  tasks_completed INTEGER DEFAULT 0,
+  tasks_failed INTEGER DEFAULT 0,
+  current_task_id INTEGER,
+  metadata JSONB DEFAULT '{}',
+  created_at TIMESTAMPTZ DEFAULT NOW(),
+  updated_at TIMESTAMPTZ DEFAULT NOW()
+);
+
+-- Indexes for worker registry
+CREATE INDEX IF NOT EXISTS idx_worker_registry_status ON worker_registry(status);
+CREATE INDEX IF NOT EXISTS idx_worker_registry_role ON worker_registry(role);
+CREATE INDEX IF NOT EXISTS idx_worker_registry_heartbeat ON worker_registry(last_heartbeat_at);
+
+-- Function to assign a name to a new worker
+CREATE OR REPLACE FUNCTION assign_worker_name(p_worker_id VARCHAR(100))
+RETURNS VARCHAR(50) AS $$
+DECLARE
+  v_name VARCHAR(50);
+BEGIN
+  -- Try to get an unused name
+  UPDATE worker_name_pool
+  SET in_use = TRUE, assigned_to = p_worker_id, assigned_at = NOW()
+  WHERE id = (
+    SELECT id FROM worker_name_pool
+    WHERE in_use = FALSE
+    ORDER BY RANDOM()
+    LIMIT 1
+    FOR UPDATE SKIP LOCKED
+  )
+  RETURNING name INTO v_name;
+
+  -- If no names available, generate one
+  IF v_name IS NULL THEN
+    v_name := 'Worker-' || SUBSTRING(p_worker_id FROM 1 FOR 8);
+  END IF;
+
+  RETURN v_name;
+END;
+$$ LANGUAGE plpgsql;
+
+-- Function to release a worker's name back to the pool
+CREATE OR REPLACE FUNCTION release_worker_name(p_worker_id VARCHAR(100))
+RETURNS VOID AS $$
+BEGIN
+  UPDATE worker_name_pool
+  SET in_use = FALSE, assigned_to = NULL, assigned_at = NULL
+  WHERE assigned_to = p_worker_id;
+END;
+$$ LANGUAGE plpgsql;
+
+-- Function to mark stale workers as offline
+CREATE OR REPLACE FUNCTION mark_stale_workers(stale_threshold_minutes INTEGER DEFAULT 5)
+RETURNS INTEGER AS $$
+DECLARE
+  v_count INTEGER;
+BEGIN
+  UPDATE worker_registry
+  SET status = 'offline', updated_at = NOW()
+  WHERE status IN ('active', 'idle', 'starting')
+    AND last_heartbeat_at < NOW() - (stale_threshold_minutes || ' minutes')::INTERVAL
+  RETURNING COUNT(*) INTO v_count;
+
+  -- Release names from offline workers
+  PERFORM release_worker_name(worker_id)
+  FROM worker_registry
+  WHERE status = 'offline'
+    AND last_heartbeat_at < NOW() - INTERVAL '30 minutes';
+
+  RETURN COALESCE(v_count, 0);
+END;
+$$ LANGUAGE plpgsql;
+
+-- View for dashboard
+CREATE OR REPLACE VIEW v_active_workers AS
+SELECT
+  wr.id,
+  wr.worker_id,
+  wr.friendly_name,
+  wr.role,
+  wr.status,
+  wr.pod_name,
+  wr.hostname,
+  wr.started_at,
+  wr.last_heartbeat_at,
+  wr.last_task_at,
+  wr.tasks_completed,
+  wr.tasks_failed,
+  wr.current_task_id,
+  EXTRACT(EPOCH FROM (NOW() - wr.last_heartbeat_at)) as seconds_since_heartbeat,
+  CASE
+    WHEN wr.status = 'offline' THEN 'offline'
+    WHEN wr.last_heartbeat_at < NOW() - INTERVAL '2 minutes' THEN 'stale'
+    WHEN wr.current_task_id IS NOT NULL THEN 'busy'
+    ELSE 'ready'
+  END as health_status
+FROM worker_registry wr
+WHERE wr.status != 'terminated'
+ORDER BY wr.status = 'active' DESC, wr.last_heartbeat_at DESC;
+
+COMMENT ON TABLE worker_registry IS 'Tracks all workers that have registered with the system';
+COMMENT ON TABLE worker_name_pool IS 'Pool of friendly names for workers - expandable via API';

backend/migrations/077_click_events_location.sql

@@ -0,0 +1,35 @@
+-- Migration: Add visitor location and dispensary name to click events
+-- Captures where visitors are clicking from and which dispensary
+
+-- Add visitor location columns
+ALTER TABLE product_click_events
+ADD COLUMN IF NOT EXISTS visitor_city VARCHAR(100);
+
+ALTER TABLE product_click_events
+ADD COLUMN IF NOT EXISTS visitor_state VARCHAR(10);
+
+ALTER TABLE product_click_events
+ADD COLUMN IF NOT EXISTS visitor_lat DECIMAL(10, 7);
+
+ALTER TABLE product_click_events
+ADD COLUMN IF NOT EXISTS visitor_lng DECIMAL(10, 7);
+
+-- Add dispensary name for easier reporting
+ALTER TABLE product_click_events
+ADD COLUMN IF NOT EXISTS dispensary_name VARCHAR(255);
+
+-- Create index for location-based analytics
+CREATE INDEX IF NOT EXISTS idx_product_click_events_visitor_state
+ON product_click_events(visitor_state)
+WHERE visitor_state IS NOT NULL;
+
+CREATE INDEX IF NOT EXISTS idx_product_click_events_visitor_city
+ON product_click_events(visitor_city)
+WHERE visitor_city IS NOT NULL;
+
+-- Add comments
+COMMENT ON COLUMN product_click_events.visitor_city IS 'City where the visitor is located (from IP geolocation)';
+COMMENT ON COLUMN product_click_events.visitor_state IS 'State where the visitor is located (from IP geolocation)';
+COMMENT ON COLUMN product_click_events.visitor_lat IS 'Visitor latitude (from IP geolocation)';
+COMMENT ON COLUMN product_click_events.visitor_lng IS 'Visitor longitude (from IP geolocation)';
+COMMENT ON COLUMN product_click_events.dispensary_name IS 'Name of the dispensary (denormalized for easier reporting)';

backend/migrations/078_proxy_consecutive_403.sql

@@ -0,0 +1,8 @@
+-- Migration 078: Add consecutive_403_count to proxies table
+-- Per workflow-12102025.md: Track consecutive 403s per proxy
+-- After 3 consecutive 403s with different fingerprints → disable proxy
+
+ALTER TABLE proxies ADD COLUMN IF NOT EXISTS consecutive_403_count INTEGER DEFAULT 0;
+
+-- Add comment explaining the column
+COMMENT ON COLUMN proxies.consecutive_403_count IS 'Tracks consecutive 403 blocks. Reset to 0 on success. Proxy disabled at 3.';

backend/migrations/079_task_schedules.sql

@@ -0,0 +1,49 @@
+-- Migration 079: Task Schedules for Database-Driven Scheduler
+-- Per TASK_WORKFLOW_2024-12-10.md: Replaces node-cron with DB-driven scheduling
+--
+-- 2024-12-10: Created for reliable, multi-replica-safe task scheduling
+
+-- task_schedules: Stores schedule definitions and state
+CREATE TABLE IF NOT EXISTS task_schedules (
+  id SERIAL PRIMARY KEY,
+  name VARCHAR(100) NOT NULL UNIQUE,
+  role VARCHAR(50) NOT NULL,  -- TaskRole: product_refresh, store_discovery, etc.
+  description TEXT,
+
+  -- Schedule configuration
+  enabled BOOLEAN DEFAULT TRUE,
+  interval_hours INTEGER NOT NULL DEFAULT 4,
+  priority INTEGER DEFAULT 0,
+
+  -- Optional scope filters
+  state_code VARCHAR(2),  -- NULL = all states
+  platform VARCHAR(50),   -- NULL = all platforms
+
+  -- Execution state (updated by scheduler)
+  last_run_at TIMESTAMPTZ,
+  next_run_at TIMESTAMPTZ,
+  last_task_count INTEGER DEFAULT 0,
+  last_error TEXT,
+
+  created_at TIMESTAMPTZ DEFAULT NOW(),
+  updated_at TIMESTAMPTZ DEFAULT NOW()
+);
+
+-- Indexes for scheduler queries
+CREATE INDEX IF NOT EXISTS idx_task_schedules_enabled ON task_schedules(enabled) WHERE enabled = TRUE;
+CREATE INDEX IF NOT EXISTS idx_task_schedules_next_run ON task_schedules(next_run_at) WHERE enabled = TRUE;
+
+-- Insert default schedules
+INSERT INTO task_schedules (name, role, interval_hours, priority, description, next_run_at)
+VALUES
+  ('product_refresh_all', 'product_refresh', 4, 0, 'Generate product refresh tasks for all crawl-enabled stores every 4 hours', NOW()),
+  ('store_discovery_dutchie', 'store_discovery', 24, 5, 'Discover new Dutchie stores daily', NOW()),
+  ('analytics_refresh', 'analytics_refresh', 6, 0, 'Refresh analytics materialized views every 6 hours', NOW())
+ON CONFLICT (name) DO NOTHING;
+
+-- Comment for documentation
+COMMENT ON TABLE task_schedules IS 'Database-driven task scheduler configuration. Per TASK_WORKFLOW_2024-12-10.md:
+- Schedules persist in DB (survive restarts)
+- Uses SELECT FOR UPDATE SKIP LOCKED for multi-replica safety
+- Scheduler polls every 60s and executes due schedules
+- Creates tasks in worker_tasks for task-worker.ts to process';

backend/migrations/080_raw_crawl_payloads.sql

@@ -0,0 +1,58 @@
+-- Migration 080: Raw Crawl Payloads Metadata Table
+-- Per TASK_WORKFLOW_2024-12-10.md: Store full GraphQL payloads for historical analysis
+--
+-- Design Pattern: Metadata/Payload Separation
+-- - Metadata (this table): Small, indexed, queryable
+-- - Payload (filesystem): Gzipped JSON at storage_path
+--
+-- Benefits:
+-- - Compare any two crawls to see what changed
+-- - Replay/re-normalize historical data if logic changes
+-- - Debug issues by seeing exactly what the API returned
+-- - DB stays small, backups stay fast
+--
+-- Storage location: /storage/payloads/{year}/{month}/{day}/store_{id}_{timestamp}.json.gz
+-- Compression: ~90% reduction (1.5MB -> 150KB per crawl)
+
+CREATE TABLE IF NOT EXISTS raw_crawl_payloads (
+  id SERIAL PRIMARY KEY,
+
+  -- Links to crawl tracking
+  crawl_run_id INTEGER REFERENCES crawl_runs(id) ON DELETE SET NULL,
+  dispensary_id INTEGER NOT NULL REFERENCES dispensaries(id) ON DELETE CASCADE,
+
+  -- File location (gzipped JSON)
+  storage_path TEXT NOT NULL,
+
+  -- Metadata for quick queries without loading file
+  product_count INTEGER NOT NULL DEFAULT 0,
+  size_bytes INTEGER,                    -- Compressed size
+  size_bytes_raw INTEGER,                -- Uncompressed size
+
+  -- Timestamps
+  fetched_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
+  created_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
+
+  -- Optional: checksum for integrity verification
+  checksum_sha256 VARCHAR(64)
+);
+
+-- Indexes for common queries
+CREATE INDEX IF NOT EXISTS idx_raw_crawl_payloads_dispensary
+  ON raw_crawl_payloads(dispensary_id);
+
+CREATE INDEX IF NOT EXISTS idx_raw_crawl_payloads_dispensary_fetched
+  ON raw_crawl_payloads(dispensary_id, fetched_at DESC);
+
+CREATE INDEX IF NOT EXISTS idx_raw_crawl_payloads_fetched
+  ON raw_crawl_payloads(fetched_at DESC);
+
+CREATE INDEX IF NOT EXISTS idx_raw_crawl_payloads_crawl_run
+  ON raw_crawl_payloads(crawl_run_id)
+  WHERE crawl_run_id IS NOT NULL;
+
+-- Comments
+COMMENT ON TABLE raw_crawl_payloads IS 'Metadata for raw GraphQL payloads stored on filesystem. Per TASK_WORKFLOW_2024-12-10.md: Full payloads enable historical diffs and replay.';
+COMMENT ON COLUMN raw_crawl_payloads.storage_path IS 'Path to gzipped JSON file, e.g. /storage/payloads/2024/12/10/store_123_1702234567.json.gz';
+COMMENT ON COLUMN raw_crawl_payloads.size_bytes IS 'Compressed file size in bytes';
+COMMENT ON COLUMN raw_crawl_payloads.size_bytes_raw IS 'Uncompressed payload size in bytes';

backend/migrations/081_payload_fetch_columns.sql

@@ -0,0 +1,37 @@
+-- Migration 081: Payload Fetch Columns
+-- Per TASK_WORKFLOW_2024-12-10.md: Separates API fetch from data processing
+--
+-- New architecture:
+-- - payload_fetch: Hits Dutchie API, saves raw payload to disk
+-- - product_refresh: Reads local payload, normalizes, upserts to DB
+--
+-- This migration adds:
+-- 1. payload column to worker_tasks (for task chaining data)
+-- 2. processed_at column to raw_crawl_payloads (track when payload was processed)
+-- 3. last_fetch_at column to dispensaries (track when last payload was fetched)
+
+-- Add payload column to worker_tasks for task chaining
+-- Used by payload_fetch to pass payload_id to product_refresh
+ALTER TABLE worker_tasks
+ADD COLUMN IF NOT EXISTS payload JSONB DEFAULT NULL;
+
+COMMENT ON COLUMN worker_tasks.payload IS 'Per TASK_WORKFLOW_2024-12-10.md: Task chaining data (e.g., payload_id from payload_fetch to product_refresh)';
+
+-- Add processed_at to raw_crawl_payloads
+-- Tracks when the payload was processed by product_refresh
+ALTER TABLE raw_crawl_payloads
+ADD COLUMN IF NOT EXISTS processed_at TIMESTAMPTZ DEFAULT NULL;
+
+COMMENT ON COLUMN raw_crawl_payloads.processed_at IS 'When this payload was processed by product_refresh handler';
+
+-- Index for finding unprocessed payloads
+CREATE INDEX IF NOT EXISTS idx_raw_crawl_payloads_unprocessed
+  ON raw_crawl_payloads(dispensary_id, fetched_at DESC)
+  WHERE processed_at IS NULL;
+
+-- Add last_fetch_at to dispensaries
+-- Tracks when the last payload was fetched (separate from last_crawl_at which is when processing completed)
+ALTER TABLE dispensaries
+ADD COLUMN IF NOT EXISTS last_fetch_at TIMESTAMPTZ DEFAULT NULL;
+
+COMMENT ON COLUMN dispensaries.last_fetch_at IS 'Per TASK_WORKFLOW_2024-12-10.md: When last payload was fetched from API (separate from last_crawl_at which is when processing completed)';

backend/migrations/082_proxy_notification_trigger.sql

@@ -0,0 +1,27 @@
+-- Migration: 082_proxy_notification_trigger
+-- Date: 2024-12-11
+-- Description: Add PostgreSQL NOTIFY trigger to alert workers when proxies are added
+
+-- Create function to notify workers when active proxy is added/activated
+CREATE OR REPLACE FUNCTION notify_proxy_added()
+RETURNS TRIGGER AS $$
+BEGIN
+  -- Only notify if proxy is active
+  IF NEW.active = true THEN
+    PERFORM pg_notify('proxy_added', NEW.id::text);
+  END IF;
+  RETURN NEW;
+END;
+$$ LANGUAGE plpgsql;
+
+-- Drop existing trigger if any
+DROP TRIGGER IF EXISTS proxy_added_trigger ON proxies;
+
+-- Create trigger on insert and update of active column
+CREATE TRIGGER proxy_added_trigger
+AFTER INSERT OR UPDATE OF active ON proxies
+FOR EACH ROW
+EXECUTE FUNCTION notify_proxy_added();
+
+COMMENT ON FUNCTION notify_proxy_added() IS
+'Sends PostgreSQL NOTIFY to proxy_added channel when an active proxy is added or activated. Workers LISTEN on this channel to wake up immediately.';

backend/migrations/083_discovery_runs.sql

@@ -0,0 +1,88 @@
+-- Migration 083: Discovery Run Tracking
+-- Tracks progress of store discovery runs step-by-step
+
+-- Main discovery runs table
+CREATE TABLE IF NOT EXISTS discovery_runs (
+  id SERIAL PRIMARY KEY,
+  platform VARCHAR(50) NOT NULL DEFAULT 'dutchie',
+  status VARCHAR(20) NOT NULL DEFAULT 'running', -- running, completed, failed
+  started_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
+  finished_at TIMESTAMPTZ,
+  task_id INTEGER REFERENCES worker_task_queue(id),
+
+  -- Totals
+  states_total INTEGER DEFAULT 0,
+  states_completed INTEGER DEFAULT 0,
+  locations_discovered INTEGER DEFAULT 0,
+  locations_promoted INTEGER DEFAULT 0,
+  new_store_ids INTEGER[] DEFAULT '{}',
+
+  -- Error info
+  error_message TEXT,
+
+  created_at TIMESTAMPTZ NOT NULL DEFAULT NOW()
+);
+
+-- Per-state progress within a run
+CREATE TABLE IF NOT EXISTS discovery_run_states (
+  id SERIAL PRIMARY KEY,
+  run_id INTEGER NOT NULL REFERENCES discovery_runs(id) ON DELETE CASCADE,
+  state_code VARCHAR(2) NOT NULL,
+  status VARCHAR(20) NOT NULL DEFAULT 'pending', -- pending, running, completed, failed
+  started_at TIMESTAMPTZ,
+  finished_at TIMESTAMPTZ,
+
+  -- Results
+  cities_found INTEGER DEFAULT 0,
+  locations_found INTEGER DEFAULT 0,
+  locations_upserted INTEGER DEFAULT 0,
+  new_dispensary_ids INTEGER[] DEFAULT '{}',
+
+  -- Error info
+  error_message TEXT,
+
+  created_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
+
+  UNIQUE(run_id, state_code)
+);
+
+-- Step-by-step log for detailed progress tracking
+CREATE TABLE IF NOT EXISTS discovery_run_steps (
+  id SERIAL PRIMARY KEY,
+  run_id INTEGER NOT NULL REFERENCES discovery_runs(id) ON DELETE CASCADE,
+  state_code VARCHAR(2),
+  step_name VARCHAR(100) NOT NULL,
+  status VARCHAR(20) NOT NULL DEFAULT 'started', -- started, completed, failed
+  started_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
+  finished_at TIMESTAMPTZ,
+
+  -- Details (JSON for flexibility)
+  details JSONB DEFAULT '{}',
+
+  created_at TIMESTAMPTZ NOT NULL DEFAULT NOW()
+);
+
+-- Indexes for querying
+CREATE INDEX IF NOT EXISTS idx_discovery_runs_status ON discovery_runs(status);
+CREATE INDEX IF NOT EXISTS idx_discovery_runs_platform ON discovery_runs(platform);
+CREATE INDEX IF NOT EXISTS idx_discovery_runs_started_at ON discovery_runs(started_at DESC);
+CREATE INDEX IF NOT EXISTS idx_discovery_run_states_run_id ON discovery_run_states(run_id);
+CREATE INDEX IF NOT EXISTS idx_discovery_run_steps_run_id ON discovery_run_steps(run_id);
+
+-- View for latest run status per platform
+CREATE OR REPLACE VIEW v_latest_discovery_runs AS
+SELECT DISTINCT ON (platform)
+  id,
+  platform,
+  status,
+  started_at,
+  finished_at,
+  states_total,
+  states_completed,
+  locations_discovered,
+  locations_promoted,
+  array_length(new_store_ids, 1) as new_stores_count,
+  error_message,
+  EXTRACT(EPOCH FROM (COALESCE(finished_at, NOW()) - started_at)) as duration_seconds
+FROM discovery_runs
+ORDER BY platform, started_at DESC;

backend/migrations/084_dual_transport_preflight.sql

@@ -0,0 +1,253 @@
+-- Migration 084: Dual Transport Preflight System
+-- Workers run both curl and http (Puppeteer) preflights on startup
+-- Tasks can require a specific transport method
+
+-- ===================================================================
+-- PART 1: Add preflight columns to worker_registry
+-- ===================================================================
+
+-- Preflight status for curl/axios transport (proxy-based)
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS preflight_curl_status VARCHAR(20) DEFAULT 'pending';
+
+-- Preflight status for http/Puppeteer transport (browser-based)
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS preflight_http_status VARCHAR(20) DEFAULT 'pending';
+
+-- Timestamps for when each preflight completed
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS preflight_curl_at TIMESTAMPTZ;
+
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS preflight_http_at TIMESTAMPTZ;
+
+-- Error messages for failed preflights
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS preflight_curl_error TEXT;
+
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS preflight_http_error TEXT;
+
+-- Response time for successful preflights (ms)
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS preflight_curl_ms INTEGER;
+
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS preflight_http_ms INTEGER;
+
+-- Constraints for preflight status values
+ALTER TABLE worker_registry
+DROP CONSTRAINT IF EXISTS valid_preflight_curl_status;
+
+ALTER TABLE worker_registry
+ADD CONSTRAINT valid_preflight_curl_status
+CHECK (preflight_curl_status IN ('pending', 'passed', 'failed', 'skipped'));
+
+ALTER TABLE worker_registry
+DROP CONSTRAINT IF EXISTS valid_preflight_http_status;
+
+ALTER TABLE worker_registry
+ADD CONSTRAINT valid_preflight_http_status
+CHECK (preflight_http_status IN ('pending', 'passed', 'failed', 'skipped'));
+
+-- ===================================================================
+-- PART 2: Add method column to worker_tasks
+-- ===================================================================
+
+-- Transport method requirement for the task
+-- NULL = no preference (any worker can claim)
+-- 'curl' = requires curl/axios transport (proxy-based, fast)
+-- 'http' = requires http/Puppeteer transport (browser-based, anti-detect)
+ALTER TABLE worker_tasks
+ADD COLUMN IF NOT EXISTS method VARCHAR(10);
+
+-- Constraint for valid method values
+ALTER TABLE worker_tasks
+DROP CONSTRAINT IF EXISTS valid_task_method;
+
+ALTER TABLE worker_tasks
+ADD CONSTRAINT valid_task_method
+CHECK (method IS NULL OR method IN ('curl', 'http'));
+
+-- Index for method-based task claiming
+CREATE INDEX IF NOT EXISTS idx_worker_tasks_method
+  ON worker_tasks(method)
+  WHERE status = 'pending';
+
+-- Set default method for all existing pending tasks to 'http'
+-- ALL current tasks require Puppeteer/browser-based transport
+UPDATE worker_tasks
+SET method = 'http'
+WHERE method IS NULL;
+
+-- ===================================================================
+-- PART 3: Update claim_task function for method compatibility
+-- ===================================================================
+
+CREATE OR REPLACE FUNCTION claim_task(
+  p_role VARCHAR(50),
+  p_worker_id VARCHAR(100),
+  p_curl_passed BOOLEAN DEFAULT TRUE,
+  p_http_passed BOOLEAN DEFAULT FALSE
+) RETURNS worker_tasks AS $$
+DECLARE
+  claimed_task worker_tasks;
+BEGIN
+  UPDATE worker_tasks
+  SET
+    status = 'claimed',
+    worker_id = p_worker_id,
+    claimed_at = NOW(),
+    updated_at = NOW()
+  WHERE id = (
+    SELECT id FROM worker_tasks
+    WHERE role = p_role
+      AND status = 'pending'
+      AND (scheduled_for IS NULL OR scheduled_for <= NOW())
+      -- Method compatibility: worker must have passed the required preflight
+      AND (
+        method IS NULL  -- No preference, any worker can claim
+        OR (method = 'curl' AND p_curl_passed = TRUE)
+        OR (method = 'http' AND p_http_passed = TRUE)
+      )
+      -- Exclude stores that already have an active task
+      AND (dispensary_id IS NULL OR dispensary_id NOT IN (
+        SELECT dispensary_id FROM worker_tasks
+        WHERE status IN ('claimed', 'running')
+        AND dispensary_id IS NOT NULL
+      ))
+    ORDER BY priority DESC, created_at ASC
+    LIMIT 1
+    FOR UPDATE SKIP LOCKED
+  )
+  RETURNING * INTO claimed_task;
+
+  RETURN claimed_task;
+END;
+$$ LANGUAGE plpgsql;
+
+-- ===================================================================
+-- PART 4: Update v_active_workers view
+-- ===================================================================
+
+DROP VIEW IF EXISTS v_active_workers;
+
+CREATE VIEW v_active_workers AS
+SELECT
+  wr.id,
+  wr.worker_id,
+  wr.friendly_name,
+  wr.role,
+  wr.status,
+  wr.pod_name,
+  wr.hostname,
+  wr.started_at,
+  wr.last_heartbeat_at,
+  wr.last_task_at,
+  wr.tasks_completed,
+  wr.tasks_failed,
+  wr.current_task_id,
+  -- Preflight status
+  wr.preflight_curl_status,
+  wr.preflight_http_status,
+  wr.preflight_curl_at,
+  wr.preflight_http_at,
+  wr.preflight_curl_error,
+  wr.preflight_http_error,
+  wr.preflight_curl_ms,
+  wr.preflight_http_ms,
+  -- Computed fields
+  EXTRACT(EPOCH FROM (NOW() - wr.last_heartbeat_at)) as seconds_since_heartbeat,
+  CASE
+    WHEN wr.status = 'offline' THEN 'offline'
+    WHEN wr.last_heartbeat_at < NOW() - INTERVAL '2 minutes' THEN 'stale'
+    WHEN wr.current_task_id IS NOT NULL THEN 'busy'
+    ELSE 'ready'
+  END as health_status,
+  -- Capability flags (can this worker handle curl/http tasks?)
+  (wr.preflight_curl_status = 'passed') as can_curl,
+  (wr.preflight_http_status = 'passed') as can_http
+FROM worker_registry wr
+WHERE wr.status != 'terminated'
+ORDER BY wr.status = 'active' DESC, wr.last_heartbeat_at DESC;
+
+-- ===================================================================
+-- PART 5: View for task queue with method info
+-- ===================================================================
+
+DROP VIEW IF EXISTS v_task_history;
+
+CREATE VIEW v_task_history AS
+SELECT
+  t.id,
+  t.role,
+  t.dispensary_id,
+  d.name as dispensary_name,
+  t.platform,
+  t.status,
+  t.priority,
+  t.method,
+  t.worker_id,
+  t.scheduled_for,
+  t.claimed_at,
+  t.started_at,
+  t.completed_at,
+  t.error_message,
+  t.retry_count,
+  t.created_at,
+  EXTRACT(EPOCH FROM (t.completed_at - t.started_at)) as duration_sec
+FROM worker_tasks t
+LEFT JOIN dispensaries d ON d.id = t.dispensary_id
+ORDER BY t.created_at DESC;
+
+-- ===================================================================
+-- PART 6: Helper function to update worker preflight status
+-- ===================================================================
+
+CREATE OR REPLACE FUNCTION update_worker_preflight(
+  p_worker_id VARCHAR(100),
+  p_transport VARCHAR(10),  -- 'curl' or 'http'
+  p_status VARCHAR(20),     -- 'passed', 'failed', 'skipped'
+  p_response_ms INTEGER DEFAULT NULL,
+  p_error TEXT DEFAULT NULL
+) RETURNS VOID AS $$
+BEGIN
+  IF p_transport = 'curl' THEN
+    UPDATE worker_registry
+    SET
+      preflight_curl_status = p_status,
+      preflight_curl_at = NOW(),
+      preflight_curl_ms = p_response_ms,
+      preflight_curl_error = p_error,
+      updated_at = NOW()
+    WHERE worker_id = p_worker_id;
+  ELSIF p_transport = 'http' THEN
+    UPDATE worker_registry
+    SET
+      preflight_http_status = p_status,
+      preflight_http_at = NOW(),
+      preflight_http_ms = p_response_ms,
+      preflight_http_error = p_error,
+      updated_at = NOW()
+    WHERE worker_id = p_worker_id;
+  END IF;
+END;
+$$ LANGUAGE plpgsql;
+
+-- ===================================================================
+-- Comments
+-- ===================================================================
+
+COMMENT ON COLUMN worker_registry.preflight_curl_status IS 'Status of curl/axios preflight: pending, passed, failed, skipped';
+COMMENT ON COLUMN worker_registry.preflight_http_status IS 'Status of http/Puppeteer preflight: pending, passed, failed, skipped';
+COMMENT ON COLUMN worker_registry.preflight_curl_at IS 'When curl preflight completed';
+COMMENT ON COLUMN worker_registry.preflight_http_at IS 'When http preflight completed';
+COMMENT ON COLUMN worker_registry.preflight_curl_error IS 'Error message if curl preflight failed';
+COMMENT ON COLUMN worker_registry.preflight_http_error IS 'Error message if http preflight failed';
+COMMENT ON COLUMN worker_registry.preflight_curl_ms IS 'Response time of successful curl preflight (ms)';
+COMMENT ON COLUMN worker_registry.preflight_http_ms IS 'Response time of successful http preflight (ms)';
+
+COMMENT ON COLUMN worker_tasks.method IS 'Transport method required: NULL=any, curl=proxy-based, http=browser-based';
+
+COMMENT ON FUNCTION claim_task IS 'Atomically claim a task, respecting method requirements and per-store locking';
+COMMENT ON FUNCTION update_worker_preflight IS 'Update a workers preflight status for a given transport';

backend/migrations/085_preflight_ip_fingerprint.sql

@@ -0,0 +1,168 @@
+-- Migration 085: Add IP and fingerprint columns for preflight reporting
+-- These columns were missing from migration 084
+
+-- ===================================================================
+-- PART 1: Add IP address columns to worker_registry
+-- ===================================================================
+
+-- IP address detected during curl/axios preflight
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS curl_ip VARCHAR(45);
+
+-- IP address detected during http/Puppeteer preflight
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS http_ip VARCHAR(45);
+
+-- ===================================================================
+-- PART 2: Add fingerprint data column
+-- ===================================================================
+
+-- Browser fingerprint data captured during Puppeteer preflight
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS fingerprint_data JSONB;
+
+-- ===================================================================
+-- PART 3: Add combined preflight status/timestamp for convenience
+-- ===================================================================
+
+-- Overall preflight status (computed from both transports)
+-- Values: 'pending', 'passed', 'partial', 'failed'
+--   - 'pending': neither transport tested
+--   - 'passed': both transports passed (or http passed for browser-only)
+--   - 'partial': at least one passed
+--   - 'failed': no transport passed
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS preflight_status VARCHAR(20) DEFAULT 'pending';
+
+-- Most recent preflight completion timestamp
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS preflight_at TIMESTAMPTZ;
+
+-- ===================================================================
+-- PART 4: Update function to set preflight status
+-- ===================================================================
+
+CREATE OR REPLACE FUNCTION update_worker_preflight(
+  p_worker_id VARCHAR(100),
+  p_transport VARCHAR(10),  -- 'curl' or 'http'
+  p_status VARCHAR(20),     -- 'passed', 'failed', 'skipped'
+  p_ip VARCHAR(45) DEFAULT NULL,
+  p_response_ms INTEGER DEFAULT NULL,
+  p_error TEXT DEFAULT NULL,
+  p_fingerprint JSONB DEFAULT NULL
+) RETURNS VOID AS $$
+DECLARE
+  v_curl_status VARCHAR(20);
+  v_http_status VARCHAR(20);
+  v_overall_status VARCHAR(20);
+BEGIN
+  IF p_transport = 'curl' THEN
+    UPDATE worker_registry
+    SET
+      preflight_curl_status = p_status,
+      preflight_curl_at = NOW(),
+      preflight_curl_ms = p_response_ms,
+      preflight_curl_error = p_error,
+      curl_ip = p_ip,
+      updated_at = NOW()
+    WHERE worker_id = p_worker_id;
+  ELSIF p_transport = 'http' THEN
+    UPDATE worker_registry
+    SET
+      preflight_http_status = p_status,
+      preflight_http_at = NOW(),
+      preflight_http_ms = p_response_ms,
+      preflight_http_error = p_error,
+      http_ip = p_ip,
+      fingerprint_data = COALESCE(p_fingerprint, fingerprint_data),
+      updated_at = NOW()
+    WHERE worker_id = p_worker_id;
+  END IF;
+
+  -- Update overall preflight status
+  SELECT preflight_curl_status, preflight_http_status
+  INTO v_curl_status, v_http_status
+  FROM worker_registry
+  WHERE worker_id = p_worker_id;
+
+  -- Compute overall status
+  IF v_curl_status = 'passed' AND v_http_status = 'passed' THEN
+    v_overall_status := 'passed';
+  ELSIF v_curl_status = 'passed' OR v_http_status = 'passed' THEN
+    v_overall_status := 'partial';
+  ELSIF v_curl_status = 'failed' OR v_http_status = 'failed' THEN
+    v_overall_status := 'failed';
+  ELSE
+    v_overall_status := 'pending';
+  END IF;
+
+  UPDATE worker_registry
+  SET
+    preflight_status = v_overall_status,
+    preflight_at = NOW()
+  WHERE worker_id = p_worker_id;
+END;
+$$ LANGUAGE plpgsql;
+
+-- ===================================================================
+-- PART 5: Update v_active_workers view
+-- ===================================================================
+
+DROP VIEW IF EXISTS v_active_workers;
+
+CREATE VIEW v_active_workers AS
+SELECT
+  wr.id,
+  wr.worker_id,
+  wr.friendly_name,
+  wr.role,
+  wr.status,
+  wr.pod_name,
+  wr.hostname,
+  wr.started_at,
+  wr.last_heartbeat_at,
+  wr.last_task_at,
+  wr.tasks_completed,
+  wr.tasks_failed,
+  wr.current_task_id,
+  -- IP addresses from preflights
+  wr.curl_ip,
+  wr.http_ip,
+  -- Combined preflight status
+  wr.preflight_status,
+  wr.preflight_at,
+  -- Detailed preflight status per transport
+  wr.preflight_curl_status,
+  wr.preflight_http_status,
+  wr.preflight_curl_at,
+  wr.preflight_http_at,
+  wr.preflight_curl_error,
+  wr.preflight_http_error,
+  wr.preflight_curl_ms,
+  wr.preflight_http_ms,
+  -- Fingerprint data
+  wr.fingerprint_data,
+  -- Computed fields
+  EXTRACT(EPOCH FROM (NOW() - wr.last_heartbeat_at)) as seconds_since_heartbeat,
+  CASE
+    WHEN wr.status = 'offline' THEN 'offline'
+    WHEN wr.last_heartbeat_at < NOW() - INTERVAL '2 minutes' THEN 'stale'
+    WHEN wr.current_task_id IS NOT NULL THEN 'busy'
+    ELSE 'ready'
+  END as health_status,
+  -- Capability flags (can this worker handle curl/http tasks?)
+  (wr.preflight_curl_status = 'passed') as can_curl,
+  (wr.preflight_http_status = 'passed') as can_http
+FROM worker_registry wr
+WHERE wr.status != 'terminated'
+ORDER BY wr.status = 'active' DESC, wr.last_heartbeat_at DESC;
+
+-- ===================================================================
+-- Comments
+-- ===================================================================
+
+COMMENT ON COLUMN worker_registry.curl_ip IS 'IP address detected during curl/axios preflight';
+COMMENT ON COLUMN worker_registry.http_ip IS 'IP address detected during Puppeteer preflight';
+COMMENT ON COLUMN worker_registry.fingerprint_data IS 'Browser fingerprint captured during Puppeteer preflight';
+COMMENT ON COLUMN worker_registry.preflight_status IS 'Overall preflight status: pending, passed, partial, failed';
+COMMENT ON COLUMN worker_registry.preflight_at IS 'Most recent preflight completion timestamp';

backend/migrations/085_trusted_origins.sql

@@ -0,0 +1,59 @@
+-- Migration 085: Trusted Origins Management
+-- Allows admin to manage trusted IPs and domains via UI instead of hardcoded values
+
+-- Trusted origins table (IPs and domains that bypass API key auth)
+CREATE TABLE IF NOT EXISTS trusted_origins (
+  id SERIAL PRIMARY KEY,
+
+  -- Origin type: 'ip', 'domain', 'pattern'
+  origin_type VARCHAR(20) NOT NULL CHECK (origin_type IN ('ip', 'domain', 'pattern')),
+
+  -- The actual value
+  -- For ip: '127.0.0.1', '::1', '192.168.1.0/24'
+  -- For domain: 'cannaiq.co', 'findadispo.com'
+  -- For pattern: '^https://.*\.cannabrands\.app$' (regex)
+  origin_value VARCHAR(255) NOT NULL,
+
+  -- Description for admin reference
+  description TEXT,
+
+  -- Active flag
+  active BOOLEAN DEFAULT true,
+
+  -- Audit
+  created_at TIMESTAMPTZ DEFAULT NOW(),
+  created_by INTEGER REFERENCES users(id),
+  updated_at TIMESTAMPTZ DEFAULT NOW(),
+
+  UNIQUE(origin_type, origin_value)
+);
+
+-- Index for quick lookups
+CREATE INDEX IF NOT EXISTS idx_trusted_origins_active ON trusted_origins(active) WHERE active = true;
+CREATE INDEX IF NOT EXISTS idx_trusted_origins_type ON trusted_origins(origin_type, active);
+
+-- Seed with current hardcoded values
+INSERT INTO trusted_origins (origin_type, origin_value, description) VALUES
+  -- Trusted IPs (localhost)
+  ('ip', '127.0.0.1', 'Localhost IPv4'),
+  ('ip', '::1', 'Localhost IPv6'),
+  ('ip', '::ffff:127.0.0.1', 'Localhost IPv4-mapped IPv6'),
+
+  -- Trusted domains
+  ('domain', 'cannaiq.co', 'CannaiQ production'),
+  ('domain', 'www.cannaiq.co', 'CannaiQ production (www)'),
+  ('domain', 'findadispo.com', 'FindADispo production'),
+  ('domain', 'www.findadispo.com', 'FindADispo production (www)'),
+  ('domain', 'findagram.co', 'Findagram production'),
+  ('domain', 'www.findagram.co', 'Findagram production (www)'),
+  ('domain', 'localhost:3010', 'Local backend dev'),
+  ('domain', 'localhost:8080', 'Local admin dev'),
+  ('domain', 'localhost:5173', 'Local Vite dev'),
+
+  -- Pattern-based (regex)
+  ('pattern', '^https://.*\.cannabrands\.app$', 'All cannabrands.app subdomains'),
+  ('pattern', '^https://.*\.cannaiq\.co$', 'All cannaiq.co subdomains')
+ON CONFLICT (origin_type, origin_value) DO NOTHING;
+
+-- Add comment
+COMMENT ON TABLE trusted_origins IS 'IPs and domains that bypass API key authentication. Managed via /admin.';

backend/migrations/086_proxy_url_column.sql

@@ -0,0 +1,10 @@
+-- Migration 086: Add proxy_url column for alternative URL formats
+-- Some proxy providers use non-standard URL formats (e.g., host:port:user:pass)
+-- This column allows storing the raw URL directly
+
+-- Add proxy_url column - if set, used directly instead of constructing from parts
+ALTER TABLE proxies
+ADD COLUMN IF NOT EXISTS proxy_url TEXT;
+
+-- Add comment
+COMMENT ON COLUMN proxies.proxy_url IS 'Raw proxy URL (if provider uses non-standard format). Takes precedence over constructed URL from host/port/user/pass.';

backend/migrations/088_discovery_payloads.sql

@@ -0,0 +1,30 @@
+-- Migration 088: Extend raw_crawl_payloads for discovery payloads
+--
+-- Enables saving raw store data from Dutchie discovery crawls.
+-- Store discovery returns raw dispensary objects - save them for historical analysis.
+
+-- Add payload_type to distinguish product crawls from discovery crawls
+ALTER TABLE raw_crawl_payloads
+ADD COLUMN IF NOT EXISTS payload_type VARCHAR(32) NOT NULL DEFAULT 'product';
+
+-- Add state_code for discovery payloads (null for product payloads)
+ALTER TABLE raw_crawl_payloads
+ADD COLUMN IF NOT EXISTS state_code VARCHAR(10);
+
+-- Add store_count for discovery payloads (alternative to product_count)
+ALTER TABLE raw_crawl_payloads
+ADD COLUMN IF NOT EXISTS store_count INTEGER;
+
+-- Make dispensary_id nullable for discovery payloads
+ALTER TABLE raw_crawl_payloads
+ALTER COLUMN dispensary_id DROP NOT NULL;
+
+-- Add index for discovery payload queries
+CREATE INDEX IF NOT EXISTS idx_raw_crawl_payloads_type_state
+  ON raw_crawl_payloads(payload_type, state_code)
+  WHERE payload_type = 'store_discovery';
+
+-- Comments
+COMMENT ON COLUMN raw_crawl_payloads.payload_type IS 'Type: product (default), store_discovery';
+COMMENT ON COLUMN raw_crawl_payloads.state_code IS 'State code for discovery payloads (e.g., AZ, MI)';
+COMMENT ON COLUMN raw_crawl_payloads.store_count IS 'Number of stores in discovery payload';

backend/migrations/089_immutable_schedules.sql

@@ -0,0 +1,105 @@
+-- Migration 089: Immutable Schedules with Per-State Product Discovery
+--
+-- Key changes:
+-- 1. Add is_immutable column - schedules can be edited but not deleted
+-- 2. Add method column - all tasks use 'http' (Puppeteer transport)
+-- 3. Store discovery weekly (168h)
+-- 4. Per-state product_discovery schedules (4h default)
+-- 5. Remove old payload_fetch schedules
+
+-- =====================================================
+-- 1) Add new columns to task_schedules
+-- =====================================================
+ALTER TABLE task_schedules
+ADD COLUMN IF NOT EXISTS is_immutable BOOLEAN DEFAULT FALSE;
+
+ALTER TABLE task_schedules
+ADD COLUMN IF NOT EXISTS method VARCHAR(10) DEFAULT 'http';
+
+-- =====================================================
+-- 2) Update store_discovery to weekly and immutable
+-- =====================================================
+UPDATE task_schedules
+SET interval_hours = 168,  -- 7 days
+    is_immutable = TRUE,
+    method = 'http',
+    description = 'Discover new Dutchie stores weekly (HTTP transport)'
+WHERE name IN ('store_discovery_dutchie', 'Store Discovery');
+
+-- Insert if doesn't exist
+INSERT INTO task_schedules (name, role, interval_hours, priority, description, is_immutable, method, platform, next_run_at)
+VALUES ('Store Discovery', 'store_discovery', 168, 5, 'Discover new Dutchie stores weekly (HTTP transport)', TRUE, 'http', 'dutchie', NOW())
+ON CONFLICT (name) DO UPDATE SET
+  interval_hours = 168,
+  is_immutable = TRUE,
+  method = 'http',
+  description = 'Discover new Dutchie stores weekly (HTTP transport)';
+
+-- =====================================================
+-- 3) Remove old payload_fetch and product_refresh_all schedules
+-- =====================================================
+DELETE FROM task_schedules WHERE name IN ('payload_fetch_all', 'product_refresh_all');
+
+-- =====================================================
+-- 4) Create per-state product_discovery schedules
+-- =====================================================
+-- One schedule per state that has dispensaries with active cannabis programs
+INSERT INTO task_schedules (name, role, state_code, interval_hours, priority, description, is_immutable, method, enabled, next_run_at)
+SELECT
+  'product_discovery_' || lower(s.code) AS name,
+  'product_discovery' AS role,
+  s.code AS state_code,
+  4 AS interval_hours,  -- 4 hours default, editable
+  10 AS priority,
+  'Product discovery for ' || s.name || ' dispensaries (HTTP transport)' AS description,
+  TRUE AS is_immutable,  -- Can edit but not delete
+  'http' AS method,
+  CASE WHEN s.is_active THEN TRUE ELSE FALSE END AS enabled,
+  -- Stagger start times: each state starts 5 minutes after the previous
+  NOW() + (ROW_NUMBER() OVER (ORDER BY s.code) * INTERVAL '5 minutes') AS next_run_at
+FROM states s
+WHERE EXISTS (
+  SELECT 1 FROM dispensaries d
+  WHERE d.state_id = s.id AND d.crawl_enabled = true
+)
+ON CONFLICT (name) DO UPDATE SET
+  is_immutable = TRUE,
+  method = 'http',
+  description = EXCLUDED.description;
+
+-- Also create schedules for states that might have stores discovered later
+INSERT INTO task_schedules (name, role, state_code, interval_hours, priority, description, is_immutable, method, enabled, next_run_at)
+SELECT
+  'product_discovery_' || lower(s.code) AS name,
+  'product_discovery' AS role,
+  s.code AS state_code,
+  4 AS interval_hours,
+  10 AS priority,
+  'Product discovery for ' || s.name || ' dispensaries (HTTP transport)' AS description,
+  TRUE AS is_immutable,
+  'http' AS method,
+  FALSE AS enabled,  -- Disabled until stores exist
+  NOW() + INTERVAL '1 hour'
+FROM states s
+WHERE NOT EXISTS (
+  SELECT 1 FROM task_schedules ts WHERE ts.name = 'product_discovery_' || lower(s.code)
+)
+ON CONFLICT (name) DO NOTHING;
+
+-- =====================================================
+-- 5) Make analytics_refresh immutable
+-- =====================================================
+UPDATE task_schedules
+SET is_immutable = TRUE, method = 'http'
+WHERE name = 'analytics_refresh';
+
+-- =====================================================
+-- 6) Add index for schedule lookups
+-- =====================================================
+CREATE INDEX IF NOT EXISTS idx_task_schedules_state_code
+  ON task_schedules(state_code)
+  WHERE state_code IS NOT NULL;
+
+-- Comments
+COMMENT ON COLUMN task_schedules.is_immutable IS 'If TRUE, schedule cannot be deleted (only edited)';
+COMMENT ON COLUMN task_schedules.method IS 'Transport method: http (Puppeteer/browser) or curl (axios)';

backend/migrations/090_modification_tracking.sql

@@ -0,0 +1,66 @@
+-- Migration 090: Add modification tracking columns
+--
+-- Tracks when records were last modified and by which task.
+-- Enables debugging, auditing, and understanding data freshness.
+--
+-- Columns added:
+--   last_modified_at      - When the record was last modified by a task
+--   last_modified_by_task - Which task role modified it (e.g., 'product_refresh')
+--   last_modified_task_id - The specific task ID that modified it
+
+-- ============================================================
+-- dispensaries table
+-- ============================================================
+ALTER TABLE dispensaries
+ADD COLUMN IF NOT EXISTS last_modified_at TIMESTAMPTZ;
+
+ALTER TABLE dispensaries
+ADD COLUMN IF NOT EXISTS last_modified_by_task VARCHAR(50);
+
+ALTER TABLE dispensaries
+ADD COLUMN IF NOT EXISTS last_modified_task_id INTEGER;
+
+-- Index for querying recently modified records
+CREATE INDEX IF NOT EXISTS idx_dispensaries_last_modified
+  ON dispensaries(last_modified_at DESC)
+  WHERE last_modified_at IS NOT NULL;
+
+-- Index for querying by task type
+CREATE INDEX IF NOT EXISTS idx_dispensaries_modified_by_task
+  ON dispensaries(last_modified_by_task)
+  WHERE last_modified_by_task IS NOT NULL;
+
+COMMENT ON COLUMN dispensaries.last_modified_at IS 'Timestamp when this record was last modified by a task';
+COMMENT ON COLUMN dispensaries.last_modified_by_task IS 'Task role that last modified this record (e.g., store_discovery_state, entry_point_discovery)';
+COMMENT ON COLUMN dispensaries.last_modified_task_id IS 'ID of the worker_tasks record that last modified this';
+
+-- ============================================================
+-- store_products table
+-- ============================================================
+ALTER TABLE store_products
+ADD COLUMN IF NOT EXISTS last_modified_at TIMESTAMPTZ;
+
+ALTER TABLE store_products
+ADD COLUMN IF NOT EXISTS last_modified_by_task VARCHAR(50);
+
+ALTER TABLE store_products
+ADD COLUMN IF NOT EXISTS last_modified_task_id INTEGER;
+
+-- Index for querying recently modified products
+CREATE INDEX IF NOT EXISTS idx_store_products_last_modified
+  ON store_products(last_modified_at DESC)
+  WHERE last_modified_at IS NOT NULL;
+
+-- Index for querying by task type
+CREATE INDEX IF NOT EXISTS idx_store_products_modified_by_task
+  ON store_products(last_modified_by_task)
+  WHERE last_modified_by_task IS NOT NULL;
+
+-- Composite index for finding products modified by a specific task
+CREATE INDEX IF NOT EXISTS idx_store_products_task_modified
+  ON store_products(dispensary_id, last_modified_at DESC)
+  WHERE last_modified_at IS NOT NULL;
+
+COMMENT ON COLUMN store_products.last_modified_at IS 'Timestamp when this record was last modified by a task';
+COMMENT ON COLUMN store_products.last_modified_by_task IS 'Task role that last modified this record (e.g., product_refresh, product_discovery)';
+COMMENT ON COLUMN store_products.last_modified_task_id IS 'ID of the worker_tasks record that last modified this';

backend/migrations/091_store_discovery_tracking.sql

@@ -0,0 +1,26 @@
+-- Migration 091: Add store discovery tracking columns
+-- Per auto-healing scheme (2025-12-12):
+-- Track when store_discovery last updated each dispensary
+-- Track when last payload was saved
+
+-- Add last_store_discovery_at to track when store_discovery updated this record
+ALTER TABLE dispensaries
+ADD COLUMN IF NOT EXISTS last_store_discovery_at TIMESTAMPTZ;
+
+-- Add last_payload_at to track when last product payload was saved
+-- (Complements last_fetch_at which tracks API fetch time)
+ALTER TABLE dispensaries
+ADD COLUMN IF NOT EXISTS last_payload_at TIMESTAMPTZ;
+
+-- Add index for finding stale discovery data
+CREATE INDEX IF NOT EXISTS idx_dispensaries_store_discovery_at
+ON dispensaries (last_store_discovery_at DESC NULLS LAST)
+WHERE crawl_enabled = true;
+
+-- Add index for finding dispensaries without recent payloads
+CREATE INDEX IF NOT EXISTS idx_dispensaries_payload_at
+ON dispensaries (last_payload_at DESC NULLS LAST)
+WHERE crawl_enabled = true;
+
+COMMENT ON COLUMN dispensaries.last_store_discovery_at IS 'When store_discovery task last updated this record';
+COMMENT ON COLUMN dispensaries.last_payload_at IS 'When last product payload was saved for this dispensary';

backend/migrations/092_fix_trulieve_urls.sql

@@ -0,0 +1,30 @@
+-- Fix 3 Trulieve/Harvest stores with incorrect menu URLs
+-- These records have NULL or mismatched platform_dispensary_id so store_discovery
+-- ON CONFLICT can't update them automatically
+
+UPDATE dispensaries
+SET
+  menu_url = 'https://dutchie.com/dispensary/svaccha-llc-nirvana-center-apache-junction',
+  updated_at = NOW()
+WHERE id = 224;
+
+UPDATE dispensaries
+SET
+  menu_url = 'https://dutchie.com/dispensary/trulieve-of-phoenix-tatum',
+  updated_at = NOW()
+WHERE id = 76;
+
+UPDATE dispensaries
+SET
+  menu_url = 'https://dutchie.com/dispensary/harvest-of-havasu',
+  updated_at = NOW()
+WHERE id = 403;
+
+-- Queue entry_point_discovery tasks to resolve their platform_dispensary_id
+-- method='http' ensures only workers that passed http preflight can claim these
+INSERT INTO worker_tasks (role, dispensary_id, priority, scheduled_for, method)
+VALUES
+  ('entry_point_discovery', 224, 5, NOW(), 'http'),
+  ('entry_point_discovery', 76, 5, NOW(), 'http'),
+  ('entry_point_discovery', 403, 5, NOW(), 'http')
+ON CONFLICT DO NOTHING;

backend/migrations/092_store_intelligence_cache.sql

@@ -0,0 +1,35 @@
+-- Migration 092: Store Intelligence Cache
+-- Pre-computed store intelligence data refreshed by analytics_refresh task
+-- Eliminates costly aggregation queries on /intelligence/stores endpoint
+
+CREATE TABLE IF NOT EXISTS store_intelligence_cache (
+  dispensary_id INTEGER PRIMARY KEY REFERENCES dispensaries(id) ON DELETE CASCADE,
+
+  -- Basic counts
+  sku_count INTEGER NOT NULL DEFAULT 0,
+  brand_count INTEGER NOT NULL DEFAULT 0,
+  snapshot_count INTEGER NOT NULL DEFAULT 0,
+
+  -- Pricing
+  avg_price_rec NUMERIC(10,2),
+  avg_price_med NUMERIC(10,2),
+  min_price NUMERIC(10,2),
+  max_price NUMERIC(10,2),
+
+  -- Category breakdown (JSONB for flexibility)
+  category_counts JSONB DEFAULT '{}',
+
+  -- Timestamps
+  last_crawl_at TIMESTAMPTZ,
+  last_refresh_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
+
+  -- Metadata
+  created_at TIMESTAMPTZ NOT NULL DEFAULT NOW()
+);
+
+-- Index for fast lookups
+CREATE INDEX IF NOT EXISTS idx_store_intelligence_cache_refresh
+ON store_intelligence_cache (last_refresh_at DESC);
+
+COMMENT ON TABLE store_intelligence_cache IS 'Pre-computed store intelligence metrics, refreshed by analytics_refresh task';
+COMMENT ON COLUMN store_intelligence_cache.category_counts IS 'JSON object mapping category_raw to product count';

backend/migrations/093_fix_mv_state_metrics.sql

@@ -0,0 +1,43 @@
+-- Migration: 093_fix_mv_state_metrics.sql
+-- Purpose: Fix mv_state_metrics to use brand_name_raw and show correct store counts
+-- Issues fixed:
+-- 1. unique_brands used brand_id (often NULL), now uses brand_name_raw
+-- 2. Added out_of_stock_products column
+-- 3. dispensary_count now correctly named
+
+-- Drop and recreate the materialized view with correct definition
+DROP MATERIALIZED VIEW IF EXISTS mv_state_metrics;
+
+CREATE MATERIALIZED VIEW mv_state_metrics AS
+SELECT
+  d.state,
+  s.name AS state_name,
+  COUNT(DISTINCT d.id) AS dispensary_count,
+  COUNT(DISTINCT CASE WHEN d.menu_type = 'dutchie' THEN d.id END) AS dutchie_stores,
+  COUNT(DISTINCT CASE WHEN d.crawl_enabled = true THEN d.id END) AS active_stores,
+  COUNT(sp.id) AS total_products,
+  COUNT(CASE WHEN COALESCE(sp.is_in_stock, true) THEN sp.id END) AS in_stock_products,
+  COUNT(CASE WHEN sp.is_in_stock = false THEN sp.id END) AS out_of_stock_products,
+  COUNT(CASE WHEN sp.is_on_special THEN sp.id END) AS on_special_products,
+  COUNT(DISTINCT sp.brand_name_raw) FILTER (WHERE sp.brand_name_raw IS NOT NULL AND sp.brand_name_raw != '') AS unique_brands,
+  COUNT(DISTINCT sp.category_raw) FILTER (WHERE sp.category_raw IS NOT NULL) AS unique_categories,
+  ROUND(AVG(sp.price_rec) FILTER (WHERE sp.price_rec > 0)::NUMERIC, 2) AS avg_price_rec,
+  MIN(sp.price_rec) FILTER (WHERE sp.price_rec > 0) AS min_price_rec,
+  MAX(sp.price_rec) FILTER (WHERE sp.price_rec > 0) AS max_price_rec,
+  NOW() AS refreshed_at
+FROM dispensaries d
+LEFT JOIN states s ON d.state = s.code
+LEFT JOIN store_products sp ON d.id = sp.dispensary_id
+WHERE d.state IS NOT NULL
+GROUP BY d.state, s.name;
+
+-- Create unique index for CONCURRENTLY refresh support
+CREATE UNIQUE INDEX idx_mv_state_metrics_state ON mv_state_metrics(state);
+
+-- Update refresh function
+CREATE OR REPLACE FUNCTION refresh_state_metrics()
+RETURNS void AS $$
+BEGIN
+  REFRESH MATERIALIZED VIEW CONCURRENTLY mv_state_metrics;
+END;
+$$ LANGUAGE plpgsql;

backend/migrations/094_import_evomi_proxies.sql

@@ -0,0 +1,516 @@
+-- Migration: Import 500 Evomi residential proxies
+-- These are sticky-session rotating proxies where password contains session ID
+-- Active is set to false - run Test All to verify and activate
+
+-- First, drop the old unique constraint that doesn't account for username/password
+ALTER TABLE proxies DROP CONSTRAINT IF EXISTS proxies_host_port_protocol_key;
+
+-- Add new unique constraint that includes username and password
+-- This allows multiple entries for the same host:port with different credentials (sessions)
+ALTER TABLE proxies ADD CONSTRAINT proxies_host_port_protocol_username_password_key
+  UNIQUE(host, port, protocol, username, password);
+
+-- Now insert all 500 proxies
+INSERT INTO proxies (host, port, protocol, username, password, active, max_connections)
+VALUES
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-4XRRPF1UQ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-5UNGX7N7K', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-9PSKYP1GU', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-GZBKKYL2S', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-YHJHM0XZU', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-ESDYQ34CJ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-GAXUMFKQI', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-2FF66K4CI', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-SUYM0R49B', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-A8VHZMEFP', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-WNRLH6NXR', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-SPSB3IUX6', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-85N76UU5Q', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-189P3LH2F', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-47DQOAGWY', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-IBT0QO7M2', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-UPXOUOH8X', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-BFQ1PH75D', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-KNTFKRY1J', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-5L8IG6DZX', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-9YE13X0BA', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-6KBHCHF0I', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-CETHHFHZ6', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-A06J8ST3I', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-YFS93P1YR', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-RB74B3R6C', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-2JW27O3EU', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-KCUX84BL0', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-1A2KSG6HO', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-4QW8ILV0E', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-0Q09GH2VL', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-16BRXBCYC', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-9W02B3R4L', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-CVAEH76YT', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-CATOG0Q5I', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-F81625L74', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-DO4AVTPK4', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-SBZPXORD5', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-JA1AWOX03', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-0FUJTRSYT', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-CM1R2RSTB', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-EHPJZCK1S', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-ZYLKORNAF', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-05A8BUD25', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-RHM1Q6O4M', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-ES5VPCE6Z', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-P0JEGLP4O', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-OC4AX88D0', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-3BN54IEBV', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-ABSC7S550', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-LNIJU6R2V', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-OYGQPPCOV', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-32YBOHQWR', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-7KGEMK4SL', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-FAW8T2EBW', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-GPV69KI9T', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-JPBHSN8M2', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-VZ1JQOF15', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-7DJXXPK1E', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-JXKQ7JVZ1', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-88Q5UQX3B', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-HAI5K0JFO', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-65SUKG0QH', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-1XFJETX1F', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-7ZNUCVCBW', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-O1DCK15LA', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-WLTEA65WB', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-KCHAFNK2P', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-6ODSZ6CUT', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-SZ8R2EFH4', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-9EPPYQREC', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-MPCBES7UI', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-FCCPL0XWZ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-GJ23UYEGI', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-RQT80689I', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-TDQO2AP5E', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-D5Q5SEUEO', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-DZN4ZTENM', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-4HVQ33VK9', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-F1HJ7GPHA', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-RM708QD2Y', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-K36N27GM5', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-O73TS0DAE', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-54QXRWEA8', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-1P6LP0365', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-WMZ2ST34E', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-175UYF58T', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-W0HTK6F28', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-D5275CTIM', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-IH2IWVZOH', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-C4VFW7GSA', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-O9XGULSNA', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-PJ1W1P5L9', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-MQQU30KPC', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-BNPIBZTYV', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-7BNRCH922', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-5AZLU117B', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-3PPJ49VJC', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-FMC8CQO74', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-VCHW23CXJ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-1S4749PCB', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-0T9DJFZPK', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-L0RMV65W3', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-FZ1ZZUQNA', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-6IFJD23DI', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-ZKUEP5XM0', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-Z8KU62CLT', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-LO77J78X1', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-27FBKYRJ4', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-0TDQTESGW', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-IMKI89WQ1', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-ANS65MIJS', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-O3T2OTT0Y', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-MWW6Z1QVM', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-TT47MX0BB', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-59CFKTM14', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-DOD61TVZN', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-RH9Y298WS', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-X98AATJ7B', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-C3UMES1W8', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-8O3J7G3PT', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-3K4OH78OJ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-N4A3JMVL1', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-HK1SRLAC9', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-Y9VLJJXVU', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-KTTH7R0EC', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-JKVX01E8T', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-HW2VPAHJO', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-7WZ9UHBH8', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-JTKFK0CP7', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-G3F27NXG5', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-K7I2JWYSP', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-CTUU8UQ0T', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-ISHMAP6RQ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-LVWNZ1LHP', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-N5CQ1YG2Z', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-XL2XY2SLZ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-UCRZVFIV1', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-VLGQFYNEL', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-YPCDM9O5Y', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-R6VA2S25E', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-4W8X8BBUL', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-5INDC8M80', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-Q8RKKOF29', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-B5ED3EFBC', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-8IC5ZXAX1', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-KCGM25D75', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-1MO06IRID', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-4QWGUGN6W', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-5T9M5KEHT', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-9KG7W7NZF', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-NYGN5R2CL', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-H61OXFCJ2', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-30WSQ4EFH', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-J36NG6MY2', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-TZU34ZA7A', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-LPWNYL74G', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-DDJTXOS4Z', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-HFOS4S185', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-2MLGIFL1M', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-CI5AHX0TC', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-WSXVCH1WN', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-8F0C3D06T', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-3YZR0664F', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-1L2VMWTM0', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-KPMCB57O7', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-N6QXQDZV3', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-35FAYFWDP', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-TVZWE2JR8', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-0WK86IKLF', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-8WBU6ESHJ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-XGU6UNM01', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-86CXNEQZC', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-NZ4LFCHE3', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-ZKB6D72RF', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-BKXNG77NS', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-3MJ332POD', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-SL9VEYNJ0', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-LY8KO43Z8', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-8KGF1XR1L', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-WT6FB54HW', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-7UQ9JMG5E', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-KX3L2040U', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-HL809F9WU', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-T9GU40ERH', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-I5O2NX3G9', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-RVOUYU3NO', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-2T3ETNUKS', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-SW0B93DZZ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-PQ55UF3K6', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-VNRWWHHJB', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-8Q26FZ7EP', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-ZWD9FA90J', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-QSGMQX3RZ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-83NZ9MEAC', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-Q9QQ4AL37', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-QBE9KD60Z', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-NRNUXUO44', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-8F0XKQ9P8', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-095JV1CJN', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-WRRSIRUTZ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-DTUD7IDQI', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-ASCEAI9LD', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-YOUM7BJZH', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-PEG2ZH9J3', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-WAUW31F78', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-GIBZ6U7AQ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-63TD9LFBG', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-0MH1N9MJB', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-YFP9RNQIK', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-SW4N5162D', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-53MWFB2MP', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-QWLUKBMIN', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-JHS6QIX9G', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-6R04HZ5UD', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-OUJLT31VN', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-6BMKW933S', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-R4GG84E4Q', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-00XAP630X', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-AK97MC2A0', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-NBS2GKGO5', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-NVFEWK4S5', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-MTV3WSYS1', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-JS8RM4JGW', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-6NL4QR1XN', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-4BUUQVSN6', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-56WEAAU3M', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-WCA56PFTF', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-TK1QAZP0B', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-SYZ5ADFXP', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-S3VLOUW6G', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-V2K1V1JWJ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-MZ6VHV5PQ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-DRZDQDPN3', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-231VVRYYA', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-06G3MC88G', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-WS52I2ZVD', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-3QTNQD55U', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-EX7ALECU3', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-DQN8TVQY6', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-FJT54OQFI', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-BLTYUF7QR', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-8DL2JXDSO', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-KBAOXIJ4Y', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-ZYL28R5UW', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-NCRDA8LYB', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-BQYKXQLXU', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-PSHCS65MR', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-90Y1WFVYZ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-4GG33NUPW', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-5Y0A79GED', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-RMZHTAD6J', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-XBSOJ5I36', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-AAJW53VNE', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-9NYSPSEL6', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-94WMY337S', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-35Y3BJQFW', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-R7WY3TMRC', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-RXAQVH0F3', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-EFQ2AVFSB', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-XPOUJSAVD', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-RSHPF5NTT', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-Z9402336V', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-OI36C5WOJ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-XEOGV1LVS', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-QIQDXG9NC', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-9IY242GGT', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-PQTEUT52E', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-18NKI3WPS', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-34U3QAA49', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-S05TYKBBF', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-B4J8WCWDD', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-HR377WC28', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-PNRR7S1T2', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-UNR0N0KJ9', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-NARQQANBE', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-8PUL1MYUU', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-KJPCT1FP3', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-XGC80N0AM', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-8Y1JN8DH3', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-Y56M31T07', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-NHYHXQSV1', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-V30RZVG7L', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-CR6V2GSOU', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-VSAF5O0LJ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-4F4BF2LFH', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-ERSMQHXNX', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-Q0TFLZQWS', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-ZXCS6SMHD', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-JHXYAUGRA', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-IT2XYWES2', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-22UCD94OG', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-VGDLQ3K35', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-O8AFL8RGX', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-9RBIZ8G9X', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-9JIU0SVBV', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-PWRBG0GWU', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-ZME1MX12T', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-A7LWRKSJP', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-5XISX0HD4', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-5T6EXKD3Z', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-10ILV351B', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-FDULBZDIY', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-SFVR6I980', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-FKV8DCZGT', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-ECRK3M3IZ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-WMKSLOF39', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-HGE60O6AL', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-RGCWDJOT8', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-DESWK5KVN', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-RD593HJ92', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-XWNCAO39B', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-AQ4XGDLX8', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-2ZOVEA1PL', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-JF4FUX83X', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-CQ228GK3B', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-XCTMU9I7U', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-M3F37T22W', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-ASZUXM9M9', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-CJVHX24WW', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-KZT4T898V', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-RI128R5TE', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-HCAG6X9MJ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-XOQENWBP7', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-1LTQGM497', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-ZLVZT4O1G', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-FTIXTXCIA', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-O2YE6QNHY', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-0JPDDBF47', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-H1FP1IFJI', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-FYBPBMY5B', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-F7BWDVC97', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-MLENB1LQ4', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-FT9YNU8UP', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-5W21Q2O5L', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-YM61QWPR3', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-XXFQJJHZM', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-H52YKCM9X', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-NT56ZNZ54', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-DRJY7BMB5', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-P6886RPXX', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-PBXW2EY5K', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-5VQCJTM36', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-NMM3GGM1J', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-1JQQ0CDSA', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-R89YI91K4', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-7L7L9MXOT', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-50Z7MXKZS', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-EGADRZTIB', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-1DR7H46H6', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-O28QZL994', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-EYTRWVERM', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-HAJZAUWJV', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-AGYO3AB89', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-V224329ZM', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-4YTMSFWYK', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-QP40RL1N1', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-CB1BVAMAH', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-9VGXUY02O', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-BCPVVKCZ3', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-VDC3CWZX7', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-7HWLI21FA', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-5QWIUJEFM', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-4C3PBMAIZ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-3QC7DM7PH', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-A6R5G3FWV', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-3A6WDE12Z', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-0F2LZA9RU', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-XGBJXMXRX', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-5YOGR8PQ1', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-LPBFBUF3N', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-TUSPGR2AY', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-G05I8M2FQ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-H5NDXJIAQ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-X8FJL8WQZ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-KIB2FQRUP', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-VNV0OYWR7', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-GKBPM3PB2', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-XVPI30KE7', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-Y3PRMJP51', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-KEPP5SBML', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-0PDUZ6QEQ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-1GHWWFLLE', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-149S2TO8O', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-1ZB6FSIGE', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-VCRQTXDZL', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-645JVC3XL', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-2HJ00JBSR', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-7FZDG2W65', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-HD6ANE3LN', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-2HS1B1J8V', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-IHOHYMDF5', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-ZYZMAFEKF', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-JO85WX5JE', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-RURJDCURW', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-FZC3BLXPJ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-B0YR2LOZ1', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-6ZFP58ZRK', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-UMZDLHQ78', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-8A2IHDXY3', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-EDYEPWUMT', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-X3TM99R12', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-DLV0UTQ72', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-SFU0ZYIM0', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-YAJ6A66NH', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-X8CFU41AU', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-CJ3Z4WP32', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-UJBLRQKXA', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-T78R8EBGH', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-DDIH55GNZ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-F1SSD4NWF', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-4BE55FKRD', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-BG2DFBL46', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-MKVMNR7W4', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-C3Z4JUGU5', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-NVP8EEEGQ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-MQFWP2LU7', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-BH873JG6H', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-3D76651SM', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-KZ7V6KWMP', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-CD8NEJFJN', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-PWXE9L30H', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-1RT95F5LR', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-Q7CEEROE5', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-Q08APOAEG', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-NNKREGLXE', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-YQEG33MKX', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-VRD9G7H5K', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-68R86GQ1G', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-BXZUKQL2M', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-QM13UD73C', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-I7OOGJLNS', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-GXDBO1IQJ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-JJZPRFMWN', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-DBTDFITGW', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-VYHL6ASIJ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-F61NNU332', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-6Z9H72KMC', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-WVOONDMA9', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-CXTSTBXN3', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-CSMZLC921', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-3FTBSARZJ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-ESHGKBXLY', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-E0YLXW5H4', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-3QFI6UMWE', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-23VOWHO88', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-02Q9U5QCH', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-3POMNSMB0', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-NTT8OWUFQ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-MT5XEHJWX', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-ILDOY0PCQ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-MN9HU4DGO', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-1YOPU7GLL', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-ZC5BM5MYB', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-UD3FXK3I9', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-LMDJOV52Y', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-N45X16BSL', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-1CBY3Z7QC', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-F0D3AO9E6', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-YQA8GUOD1', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-2EE999233', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-D6GD5WT2Y', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-7DFBMLTMY', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-J6TJKC6VJ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-2AWQ3ZRF4', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-4KOVIF5W3', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-3489SXI1U', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-F37VKUHVE', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-GHBMAVCE4', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-W64U46547', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-1GUJV1MGQ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-M13IOZVI9', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-TX7EVZN1Z', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-2PTS2ML8J', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-VTG83RVX7', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-2IOE6BR66', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-I68XZMR23', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-Q940UN6MU', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-8Y9NFR0N0', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-MYP341DZ8', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-WJ68VGKAZ', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-819MSDR9H', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-27CGND4VG', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-YYDOD47BF', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-YU7F6J8G5', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-HMY16WTCA', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-FPWEBRLG2', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-FGE79X0DE', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-551LMZ84R', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-UWMBDCTX4', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-BNHQXW9HY', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-WB0P5LCN6', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-Z4P9E1SVG', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-UVW2G9IRN', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-OO93WVLB0', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-NTRIK82TG', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-8TXV42S74', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-Z74LKL50G', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-QQEXNIPTR', false, 1),
+('core-residential.evomi.com', 1000, 'http', 'kl8', 'ogh9U1Xe7Gzxzozo4rmP_country-US_session-WGK2VD34L', false, 1)
+ON CONFLICT DO NOTHING;

backend/migrations/095_proxy_auto_retry.sql

@@ -0,0 +1,81 @@
+-- Migration: Auto-retry failed proxies after cooldown period
+-- Proxies that fail will be retried after a configurable interval
+
+-- Add last_failed_at column to track when proxy last failed
+ALTER TABLE proxies ADD COLUMN IF NOT EXISTS last_failed_at TIMESTAMP;
+
+-- Add retry settings
+INSERT INTO settings (key, value, description)
+VALUES
+  ('proxy_retry_interval_hours', '4', 'Hours to wait before retrying a failed proxy'),
+  ('proxy_max_failures_before_permanent', '10', 'Max failures before proxy is permanently disabled')
+ON CONFLICT (key) DO NOTHING;
+
+-- Create function to get eligible proxies (active OR failed but past retry interval)
+CREATE OR REPLACE FUNCTION get_eligible_proxy_ids()
+RETURNS TABLE(proxy_id INT) AS $$
+DECLARE
+  retry_hours INT;
+BEGIN
+  -- Get retry interval from settings (default 4 hours)
+  SELECT COALESCE(value::int, 4) INTO retry_hours
+  FROM settings WHERE key = 'proxy_retry_interval_hours';
+
+  RETURN QUERY
+  SELECT p.id
+  FROM proxies p
+  WHERE p.active = true
+     OR (
+       p.active = false
+       AND p.last_failed_at IS NOT NULL
+       AND p.last_failed_at < NOW() - (retry_hours || ' hours')::interval
+       AND p.failure_count < 10  -- Don't retry if too many failures
+     )
+  ORDER BY
+    p.active DESC,  -- Prefer active proxies
+    p.failure_count ASC,  -- Then prefer proxies with fewer failures
+    RANDOM();
+END;
+$$ LANGUAGE plpgsql;
+
+-- Create scheduled job to periodically re-enable proxies past their retry window
+-- This runs every hour and marks proxies as active if they're past retry interval
+CREATE OR REPLACE FUNCTION auto_reenable_proxies()
+RETURNS INT AS $$
+DECLARE
+  retry_hours INT;
+  max_failures INT;
+  reenabled_count INT;
+BEGIN
+  -- Get settings
+  SELECT COALESCE(value::int, 4) INTO retry_hours
+  FROM settings WHERE key = 'proxy_retry_interval_hours';
+
+  SELECT COALESCE(value::int, 10) INTO max_failures
+  FROM settings WHERE key = 'proxy_max_failures_before_permanent';
+
+  -- Re-enable proxies that have cooled down
+  UPDATE proxies
+  SET active = true,
+      updated_at = NOW()
+  WHERE active = false
+    AND last_failed_at IS NOT NULL
+    AND last_failed_at < NOW() - (retry_hours || ' hours')::interval
+    AND failure_count < max_failures;
+
+  GET DIAGNOSTICS reenabled_count = ROW_COUNT;
+
+  IF reenabled_count > 0 THEN
+    RAISE NOTICE 'Auto-reenabled % proxies after % hour cooldown', reenabled_count, retry_hours;
+  END IF;
+
+  RETURN reenabled_count;
+END;
+$$ LANGUAGE plpgsql;
+
+-- Add index for efficient querying
+CREATE INDEX IF NOT EXISTS idx_proxies_retry
+ON proxies(active, last_failed_at, failure_count);
+
+COMMENT ON COLUMN proxies.last_failed_at IS 'Timestamp of last failure - used for auto-retry logic';
+COMMENT ON FUNCTION auto_reenable_proxies() IS 'Call periodically to re-enable failed proxies that have cooled down';

backend/migrations/096_product_search_indexes.sql

@@ -0,0 +1,20 @@
+-- Migration: Add trigram indexes for fast ILIKE product searches
+-- Enables fast searches on name_raw, brand_name_raw, and description
+
+-- Enable pg_trgm extension if not already enabled
+CREATE EXTENSION IF NOT EXISTS pg_trgm;
+
+-- Create GIN trigram indexes for fast ILIKE searches
+CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_store_products_name_trgm
+ON store_products USING gin (name_raw gin_trgm_ops);
+
+CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_store_products_brand_name_trgm
+ON store_products USING gin (brand_name_raw gin_trgm_ops);
+
+CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_store_products_description_trgm
+ON store_products USING gin (description gin_trgm_ops);
+
+-- Add comment
+COMMENT ON INDEX idx_store_products_name_trgm IS 'Trigram index for fast ILIKE searches on product name';
+COMMENT ON INDEX idx_store_products_brand_name_trgm IS 'Trigram index for fast ILIKE searches on brand name';
+COMMENT ON INDEX idx_store_products_description_trgm IS 'Trigram index for fast ILIKE searches on description';

backend/migrations/097_worker_tasks_worker_id_index.sql

@@ -0,0 +1,11 @@
+-- Migration: Add indexes for dashboard performance
+-- Speeds up the tasks listing query with ORDER BY and JOIN
+
+-- Index for JOIN with worker_registry
+CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_worker_tasks_worker_id
+ON worker_tasks(worker_id)
+WHERE worker_id IS NOT NULL;
+
+-- Index for ORDER BY created_at DESC (dashboard listing)
+CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_worker_tasks_created_at_desc
+ON worker_tasks(created_at DESC);

backend/migrations/098_dispensary_stage_tracking.sql

@@ -0,0 +1,13 @@
+-- Migration: Add stage tracking columns to dispensaries table
+-- Required for stage checkpoint feature in task handlers
+
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS consecutive_successes INTEGER DEFAULT 0;
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS consecutive_failures INTEGER DEFAULT 0;
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS last_successful_crawl_at TIMESTAMPTZ;
+
+-- Indexes for finding stores by status
+CREATE INDEX IF NOT EXISTS idx_dispensaries_consecutive_successes
+ON dispensaries(consecutive_successes) WHERE consecutive_successes > 0;
+
+CREATE INDEX IF NOT EXISTS idx_dispensaries_consecutive_failures
+ON dispensaries(consecutive_failures) WHERE consecutive_failures > 0;

backend/migrations/099_working_hours.sql

@@ -0,0 +1,68 @@
+-- Migration: 099_working_hours.sql
+-- Description: Working hours profiles for natural traffic pattern simulation
+-- Created: 2024-12-13
+
+-- Working hours table: defines hourly activity weights to mimic natural traffic
+CREATE TABLE IF NOT EXISTS working_hours (
+  id SERIAL PRIMARY KEY,
+  name VARCHAR(50) UNIQUE NOT NULL,
+  description TEXT,
+
+  -- Hour weights: {"0": 15, "1": 5, ..., "18": 100, ...}
+  -- Value = percent chance to trigger activity that hour (0-100)
+  hour_weights JSONB NOT NULL,
+
+  -- Day-of-week multipliers (0=Sunday, 6=Saturday)
+  -- Optional adjustment for weekend vs weekday patterns
+  dow_weights JSONB DEFAULT '{"0": 90, "1": 100, "2": 100, "3": 100, "4": 100, "5": 110, "6": 95}',
+
+  timezone VARCHAR(50) DEFAULT 'America/Phoenix',
+  enabled BOOLEAN DEFAULT true,
+  created_at TIMESTAMPTZ DEFAULT NOW(),
+  updated_at TIMESTAMPTZ DEFAULT NOW()
+);
+
+-- Seed: Natural traffic pattern based on internet usage research
+-- Optimized for cannabis dispensary browsing (lunch + after-work peaks)
+INSERT INTO working_hours (name, description, timezone, hour_weights) VALUES (
+  'natural_traffic',
+  'Mimics natural user browsing patterns - peaks at lunch and 5-7 PM',
+  'America/Phoenix',
+  '{
+    "0": 15,
+    "1": 5,
+    "2": 5,
+    "3": 5,
+    "4": 5,
+    "5": 10,
+    "6": 20,
+    "7": 30,
+    "8": 35,
+    "9": 45,
+    "10": 50,
+    "11": 60,
+    "12": 75,
+    "13": 65,
+    "14": 60,
+    "15": 70,
+    "16": 80,
+    "17": 95,
+    "18": 100,
+    "19": 100,
+    "20": 90,
+    "21": 70,
+    "22": 45,
+    "23": 25
+  }'::jsonb
+) ON CONFLICT (name) DO UPDATE SET
+  hour_weights = EXCLUDED.hour_weights,
+  description = EXCLUDED.description,
+  updated_at = NOW();
+
+-- Index for quick lookups
+CREATE INDEX IF NOT EXISTS idx_working_hours_name ON working_hours(name);
+CREATE INDEX IF NOT EXISTS idx_working_hours_enabled ON working_hours(enabled);
+
+COMMENT ON TABLE working_hours IS 'Activity profiles for natural traffic simulation. Hour weights are percent chance (0-100) to trigger activity.';
+COMMENT ON COLUMN working_hours.hour_weights IS 'JSON object mapping hour (0-23) to percent chance (0-100). 100 = always run, 0 = never run.';
+COMMENT ON COLUMN working_hours.dow_weights IS 'Optional day-of-week multipliers. 0=Sunday. Applied as (hour_weight * dow_weight / 100).';

backend/migrations/100_worker_timezone.sql

@@ -0,0 +1,19 @@
+-- Migration: 100_worker_timezone.sql
+-- Description: Add timezone column to worker_registry for working hours support
+-- Created: 2024-12-13
+
+-- Add timezone column to worker_registry
+-- Populated from preflight IP geolocation (e.g., 'America/New_York')
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS timezone VARCHAR(50);
+
+-- Add working_hours_id to link worker to a specific working hours profile
+-- NULL means use default 'natural_traffic' profile
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS working_hours_id INTEGER REFERENCES working_hours(id);
+
+-- Index for workers by timezone (useful for capacity planning)
+CREATE INDEX IF NOT EXISTS idx_worker_registry_timezone ON worker_registry(timezone);
+
+COMMENT ON COLUMN worker_registry.timezone IS 'IANA timezone from preflight IP geolocation (e.g., America/New_York)';
+COMMENT ON COLUMN worker_registry.working_hours_id IS 'Reference to working_hours profile. NULL uses default natural_traffic.';

backend/migrations/101_worker_preflight_timezone.sql

@@ -0,0 +1,78 @@
+-- Migration: 101_worker_preflight_timezone.sql
+-- Description: Update update_worker_preflight to extract timezone from fingerprint
+-- Created: 2024-12-13
+
+CREATE OR REPLACE FUNCTION public.update_worker_preflight(
+  p_worker_id character varying,
+  p_transport character varying,
+  p_status character varying,
+  p_ip character varying DEFAULT NULL,
+  p_response_ms integer DEFAULT NULL,
+  p_error text DEFAULT NULL,
+  p_fingerprint jsonb DEFAULT NULL
+)
+RETURNS void
+LANGUAGE plpgsql
+AS $function$
+DECLARE
+  v_curl_status VARCHAR(20);
+  v_http_status VARCHAR(20);
+  v_overall_status VARCHAR(20);
+  v_timezone VARCHAR(50);
+BEGIN
+  IF p_transport = 'curl' THEN
+    UPDATE worker_registry
+    SET
+      preflight_curl_status = p_status,
+      preflight_curl_at = NOW(),
+      preflight_curl_ms = p_response_ms,
+      preflight_curl_error = p_error,
+      curl_ip = p_ip,
+      updated_at = NOW()
+    WHERE worker_id = p_worker_id;
+
+  ELSIF p_transport = 'http' THEN
+    -- Extract timezone from fingerprint JSON if present
+    v_timezone := p_fingerprint->>'detectedTimezone';
+
+    UPDATE worker_registry
+    SET
+      preflight_http_status = p_status,
+      preflight_http_at = NOW(),
+      preflight_http_ms = p_response_ms,
+      preflight_http_error = p_error,
+      http_ip = p_ip,
+      fingerprint_data = COALESCE(p_fingerprint, fingerprint_data),
+      -- Save extracted timezone
+      timezone = COALESCE(v_timezone, timezone),
+      updated_at = NOW()
+    WHERE worker_id = p_worker_id;
+  END IF;
+
+  -- Update overall preflight status
+  SELECT preflight_curl_status, preflight_http_status
+  INTO v_curl_status, v_http_status
+  FROM worker_registry
+  WHERE worker_id = p_worker_id;
+
+  -- Compute overall status
+  IF v_curl_status = 'passed' AND v_http_status = 'passed' THEN
+    v_overall_status := 'passed';
+  ELSIF v_curl_status = 'passed' OR v_http_status = 'passed' THEN
+    v_overall_status := 'partial';
+  ELSIF v_curl_status = 'failed' OR v_http_status = 'failed' THEN
+    v_overall_status := 'failed';
+  ELSE
+    v_overall_status := 'pending';
+  END IF;
+
+  UPDATE worker_registry
+  SET
+    preflight_status = v_overall_status,
+    preflight_at = NOW()
+  WHERE worker_id = p_worker_id;
+END;
+$function$;
+
+COMMENT ON FUNCTION update_worker_preflight(varchar, varchar, varchar, varchar, integer, text, jsonb)
+  IS 'Updates worker preflight status and extracts timezone from fingerprint for working hours';

backend/migrations/102_check_working_hours.sql

@@ -0,0 +1,114 @@
+-- Migration: 102_check_working_hours.sql
+-- Description: Function to check if worker should be available based on working hours
+-- Created: 2024-12-13
+
+-- Function to check if a worker should be available for work
+-- Returns TRUE if worker passes the probability check for current hour
+-- Returns FALSE if worker should sleep/skip this cycle
+CREATE OR REPLACE FUNCTION check_working_hours(
+  p_worker_id VARCHAR,
+  p_profile_name VARCHAR DEFAULT 'natural_traffic'
+)
+RETURNS TABLE (
+  is_available BOOLEAN,
+  current_hour INTEGER,
+  hour_weight INTEGER,
+  worker_timezone VARCHAR,
+  roll INTEGER,
+  reason TEXT
+)
+LANGUAGE plpgsql
+AS $function$
+DECLARE
+  v_timezone VARCHAR(50);
+  v_hour INTEGER;
+  v_weight INTEGER;
+  v_dow INTEGER;
+  v_dow_weight INTEGER;
+  v_final_weight INTEGER;
+  v_roll INTEGER;
+  v_hour_weights JSONB;
+  v_dow_weights JSONB;
+  v_profile_enabled BOOLEAN;
+BEGIN
+  -- Get worker's timezone (from preflight)
+  SELECT wr.timezone INTO v_timezone
+  FROM worker_registry wr
+  WHERE wr.worker_id = p_worker_id;
+
+  -- Default to America/Phoenix if no timezone set
+  v_timezone := COALESCE(v_timezone, 'America/Phoenix');
+
+  -- Get current hour in worker's timezone
+  v_hour := EXTRACT(HOUR FROM NOW() AT TIME ZONE v_timezone)::INTEGER;
+
+  -- Get day of week (0=Sunday)
+  v_dow := EXTRACT(DOW FROM NOW() AT TIME ZONE v_timezone)::INTEGER;
+
+  -- Get working hours profile
+  SELECT wh.hour_weights, wh.dow_weights, wh.enabled
+  INTO v_hour_weights, v_dow_weights, v_profile_enabled
+  FROM working_hours wh
+  WHERE wh.name = p_profile_name AND wh.enabled = true;
+
+  -- If profile not found or disabled, always available
+  IF v_hour_weights IS NULL THEN
+    RETURN QUERY SELECT
+      TRUE::BOOLEAN,
+      v_hour,
+      100::INTEGER,
+      v_timezone,
+      0::INTEGER,
+      'Profile not found or disabled - defaulting to available'::TEXT;
+    RETURN;
+  END IF;
+
+  -- Get hour weight (default to 50 if hour not specified)
+  v_weight := COALESCE((v_hour_weights->>v_hour::TEXT)::INTEGER, 50);
+
+  -- Get day-of-week weight (default to 100)
+  v_dow_weight := COALESCE((v_dow_weights->>v_dow::TEXT)::INTEGER, 100);
+
+  -- Calculate final weight (hour_weight * dow_weight / 100)
+  v_final_weight := (v_weight * v_dow_weight / 100);
+
+  -- Roll the dice (0-99)
+  v_roll := floor(random() * 100)::INTEGER;
+
+  -- Return result
+  RETURN QUERY SELECT
+    (v_roll < v_final_weight)::BOOLEAN AS is_available,
+    v_hour AS current_hour,
+    v_final_weight AS hour_weight,
+    v_timezone AS worker_timezone,
+    v_roll AS roll,
+    CASE
+      WHEN v_roll < v_final_weight THEN
+        format('Available: rolled %s < %s%% threshold', v_roll, v_final_weight)
+      ELSE
+        format('Sleeping: rolled %s >= %s%% threshold', v_roll, v_final_weight)
+    END AS reason;
+END;
+$function$;
+
+-- Simplified version that just returns boolean
+CREATE OR REPLACE FUNCTION is_worker_available(
+  p_worker_id VARCHAR,
+  p_profile_name VARCHAR DEFAULT 'natural_traffic'
+)
+RETURNS BOOLEAN
+LANGUAGE plpgsql
+AS $function$
+DECLARE
+  v_result BOOLEAN;
+BEGIN
+  SELECT is_available INTO v_result
+  FROM check_working_hours(p_worker_id, p_profile_name);
+  RETURN COALESCE(v_result, TRUE);
+END;
+$function$;
+
+COMMENT ON FUNCTION check_working_hours(VARCHAR, VARCHAR) IS
+  'Check if worker should be available based on working hours profile. Returns detailed info.';
+COMMENT ON FUNCTION is_worker_available(VARCHAR, VARCHAR) IS
+  'Simple boolean check if worker passes working hours probability roll.';

backend/migrations/103_schedule_dispensary_id.sql

@@ -0,0 +1,12 @@
+-- Migration: 103_schedule_dispensary_id.sql
+-- Description: Add dispensary_id to task_schedules for per-store schedules
+-- Created: 2025-12-13
+
+-- Add dispensary_id column for single-store schedules
+ALTER TABLE task_schedules
+ADD COLUMN IF NOT EXISTS dispensary_id INTEGER REFERENCES dispensaries(id);
+
+-- Index for quick lookups
+CREATE INDEX IF NOT EXISTS idx_task_schedules_dispensary_id ON task_schedules(dispensary_id);
+
+COMMENT ON COLUMN task_schedules.dispensary_id IS 'For single-store schedules. If set, only this store is refreshed. If NULL, uses state_code for all stores in state.';

backend/migrations/104_task_source_tracking.sql

@@ -0,0 +1,25 @@
+-- Migration 104: Add source tracking to worker_tasks
+-- Purpose: Track WHERE tasks are created from (schedule vs API endpoint)
+--
+-- All automated task creation should be visible in task_schedules.
+-- This column helps identify "phantom" tasks created outside the schedule system.
+
+-- Add source column to worker_tasks
+ALTER TABLE worker_tasks
+ADD COLUMN IF NOT EXISTS source VARCHAR(100);
+
+-- Add source_id column (references schedule_id if from a schedule)
+ALTER TABLE worker_tasks
+ADD COLUMN IF NOT EXISTS source_schedule_id INTEGER REFERENCES task_schedules(id);
+
+-- Add request metadata (IP, user agent) for debugging
+ALTER TABLE worker_tasks
+ADD COLUMN IF NOT EXISTS source_metadata JSONB;
+
+-- Create index for querying by source
+CREATE INDEX IF NOT EXISTS idx_worker_tasks_source ON worker_tasks(source);
+
+-- Comment explaining source values
+COMMENT ON COLUMN worker_tasks.source IS 'Task creation source: schedule, api_run_now, api_crawl_state, api_batch_staggered, api_batch_az_stores, task_chain, manual';
+COMMENT ON COLUMN worker_tasks.source_schedule_id IS 'ID of the schedule that created this task (if source=schedule or source=api_run_now)';
+COMMENT ON COLUMN worker_tasks.source_metadata IS 'Request metadata: {ip, user_agent, endpoint, timestamp}';

backend/migrations/105_dashboard_performance_indexes.sql

@@ -0,0 +1,25 @@
+-- Migration 105: Add indexes for dashboard performance
+-- Purpose: Speed up the /dashboard and /national/summary endpoints
+--
+-- These queries were identified as slow:
+-- 1. COUNT(*) FROM store_product_snapshots WHERE captured_at >= NOW() - INTERVAL '24 hours'
+-- 2. National summary aggregate queries
+
+-- Index for snapshot counts by time (used in dashboard)
+CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_store_product_snapshots_captured_at
+ON store_product_snapshots(captured_at DESC);
+
+-- Index for crawl traces by time and success (used in dashboard)
+CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_crawl_traces_started_success
+ON crawl_orchestration_traces(started_at DESC, success);
+
+-- Partial index for recent failed crawls (faster for dashboard alerts)
+CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_crawl_traces_recent_failures
+ON crawl_orchestration_traces(started_at DESC)
+WHERE success = false;
+
+-- Composite index for store_products aggregations by dispensary
+-- Helps with national summary state metrics query
+CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_store_products_dispensary_brand
+ON store_products(dispensary_id, brand_name_raw)
+WHERE brand_name_raw IS NOT NULL;

backend/migrations/106_rename_store_discovery_schedule.sql

@@ -0,0 +1,10 @@
+-- Migration: 106_rename_store_discovery_schedule.sql
+-- Description: Rename store_discovery_dutchie to 'Store Discovery'
+-- Created: 2025-12-13
+
+-- Update the schedule name for better display
+-- The platform='dutchie' field is preserved for badge display in UI
+UPDATE task_schedules
+SET name = 'Store Discovery',
+    updated_at = NOW()
+WHERE name = 'store_discovery_dutchie';

backend/migrations/107_proxy_tracking.sql

@@ -0,0 +1,23 @@
+-- Migration: 107_proxy_tracking.sql
+-- Description: Add proxy tracking columns to worker_tasks for geo-targeting visibility
+-- Created: 2025-12-13
+
+-- Add proxy tracking columns to worker_tasks
+ALTER TABLE worker_tasks
+ADD COLUMN IF NOT EXISTS proxy_ip VARCHAR(45);
+
+ALTER TABLE worker_tasks
+ADD COLUMN IF NOT EXISTS proxy_geo VARCHAR(100);
+
+ALTER TABLE worker_tasks
+ADD COLUMN IF NOT EXISTS proxy_source VARCHAR(10);
+
+-- Comments
+COMMENT ON COLUMN worker_tasks.proxy_ip IS 'IP address of proxy used for this task';
+COMMENT ON COLUMN worker_tasks.proxy_geo IS 'Geo target used (e.g., "arizona", "phoenix, arizona")';
+COMMENT ON COLUMN worker_tasks.proxy_source IS 'Source of proxy: "api" (Evomi dynamic) or "static" (fallback table)';
+
+-- Index for proxy analysis
+CREATE INDEX IF NOT EXISTS idx_worker_tasks_proxy_ip
+  ON worker_tasks(proxy_ip)
+  WHERE proxy_ip IS NOT NULL;

backend/migrations/108_worker_geo_sessions.sql

@@ -0,0 +1,231 @@
+-- Migration: 108_worker_geo_sessions.sql
+-- Description: Add geo session tracking to worker_registry for state-based task assignment
+-- Created: 2025-12-13
+
+-- Worker geo session columns
+-- Worker qualifies with a geo (state/city), then only claims tasks matching that geo
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS current_state VARCHAR(2);
+
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS current_city VARCHAR(100);
+
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS geo_session_started_at TIMESTAMPTZ;
+
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS session_task_count INT DEFAULT 0;
+
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS session_max_tasks INT DEFAULT 7;
+
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS proxy_geo VARCHAR(100);
+
+-- Comments
+COMMENT ON COLUMN worker_registry.current_state IS 'Worker''s current geo assignment (US state code, e.g., AZ)';
+COMMENT ON COLUMN worker_registry.current_city IS 'Worker''s current city assignment (optional, e.g., phoenix)';
+COMMENT ON COLUMN worker_registry.geo_session_started_at IS 'When worker''s current geo session started';
+COMMENT ON COLUMN worker_registry.session_task_count IS 'Number of tasks completed in current geo session';
+COMMENT ON COLUMN worker_registry.session_max_tasks IS 'Max tasks per geo session before re-qualification (default 7)';
+COMMENT ON COLUMN worker_registry.proxy_geo IS 'Geo target string used for proxy (e.g., "arizona" or "phoenix, arizona")';
+
+-- Index for finding workers by state
+CREATE INDEX IF NOT EXISTS idx_worker_registry_current_state
+  ON worker_registry(current_state)
+  WHERE current_state IS NOT NULL;
+
+-- ============================================================
+-- UPDATED claim_task FUNCTION
+-- Now filters by worker's geo session state
+-- ============================================================
+CREATE OR REPLACE FUNCTION claim_task(
+  p_role VARCHAR(50),
+  p_worker_id VARCHAR(100),
+  p_curl_passed BOOLEAN DEFAULT TRUE,
+  p_http_passed BOOLEAN DEFAULT FALSE
+) RETURNS worker_tasks AS $$
+DECLARE
+  claimed_task worker_tasks;
+  worker_state VARCHAR(2);
+  session_valid BOOLEAN;
+  session_tasks INT;
+  max_tasks INT;
+BEGIN
+  -- Get worker's current geo session info
+  SELECT
+    current_state,
+    session_task_count,
+    session_max_tasks,
+    (geo_session_started_at IS NOT NULL AND geo_session_started_at > NOW() - INTERVAL '60 minutes')
+  INTO worker_state, session_tasks, max_tasks, session_valid
+  FROM worker_registry
+  WHERE worker_id = p_worker_id;
+
+  -- If no valid geo session, or session exhausted, worker can't claim tasks
+  -- Worker must re-qualify first
+  IF worker_state IS NULL OR NOT session_valid OR session_tasks >= COALESCE(max_tasks, 7) THEN
+    RETURN NULL;
+  END IF;
+
+  -- Claim task matching worker's state
+  UPDATE worker_tasks
+  SET
+    status = 'claimed',
+    worker_id = p_worker_id,
+    claimed_at = NOW(),
+    updated_at = NOW()
+  WHERE id = (
+    SELECT wt.id FROM worker_tasks wt
+    JOIN dispensaries d ON wt.dispensary_id = d.id
+    WHERE wt.role = p_role
+      AND wt.status = 'pending'
+      AND (wt.scheduled_for IS NULL OR wt.scheduled_for <= NOW())
+      -- GEO FILTER: Task's dispensary must match worker's state
+      AND d.state = worker_state
+      -- Method compatibility: worker must have passed the required preflight
+      AND (
+        wt.method IS NULL  -- No preference, any worker can claim
+        OR (wt.method = 'curl' AND p_curl_passed = TRUE)
+        OR (wt.method = 'http' AND p_http_passed = TRUE)
+      )
+      -- Exclude stores that already have an active task
+      AND (wt.dispensary_id IS NULL OR wt.dispensary_id NOT IN (
+        SELECT dispensary_id FROM worker_tasks
+        WHERE status IN ('claimed', 'running')
+        AND dispensary_id IS NOT NULL
+      ))
+    ORDER BY wt.priority DESC, wt.created_at ASC
+    LIMIT 1
+    FOR UPDATE SKIP LOCKED
+  )
+  RETURNING * INTO claimed_task;
+
+  -- If task claimed, increment session task count
+  -- Note: Use claimed_task.id IS NOT NULL (not claimed_task IS NOT NULL)
+  -- PostgreSQL composite type NULL check quirk
+  IF claimed_task.id IS NOT NULL THEN
+    UPDATE worker_registry
+    SET session_task_count = session_task_count + 1
+    WHERE worker_id = p_worker_id;
+  END IF;
+
+  RETURN claimed_task;
+END;
+$$ LANGUAGE plpgsql;
+
+-- ============================================================
+-- FUNCTION: assign_worker_geo
+-- Assigns a geo session to a worker based on demand
+-- Returns the assigned state, or NULL if no tasks available
+-- ============================================================
+CREATE OR REPLACE FUNCTION assign_worker_geo(
+  p_worker_id VARCHAR(100)
+) RETURNS VARCHAR(2) AS $$
+DECLARE
+  assigned_state VARCHAR(2);
+BEGIN
+  -- Find state with highest demand (pending tasks) and lowest coverage (workers)
+  SELECT d.state INTO assigned_state
+  FROM dispensaries d
+  JOIN worker_tasks wt ON wt.dispensary_id = d.id
+  LEFT JOIN worker_registry wr ON wr.current_state = d.state
+    AND wr.status = 'active'
+    AND wr.geo_session_started_at > NOW() - INTERVAL '60 minutes'
+  WHERE wt.status = 'pending'
+    AND d.platform_dispensary_id IS NOT NULL
+  GROUP BY d.state
+  ORDER BY
+    COUNT(wt.id) DESC,  -- Most pending tasks first
+    COUNT(DISTINCT wr.worker_id) ASC  -- Fewest workers second
+  LIMIT 1;
+
+  -- If no pending tasks anywhere, return NULL
+  IF assigned_state IS NULL THEN
+    RETURN NULL;
+  END IF;
+
+  -- Assign the state to this worker
+  UPDATE worker_registry
+  SET
+    current_state = assigned_state,
+    current_city = NULL,  -- City assigned later if available
+    geo_session_started_at = NOW(),
+    session_task_count = 0
+  WHERE worker_id = p_worker_id;
+
+  RETURN assigned_state;
+END;
+$$ LANGUAGE plpgsql;
+
+-- ============================================================
+-- FUNCTION: check_worker_geo_session
+-- Returns info about worker's current geo session
+-- ============================================================
+CREATE OR REPLACE FUNCTION check_worker_geo_session(
+  p_worker_id VARCHAR(100)
+) RETURNS TABLE (
+  current_state VARCHAR(2),
+  current_city VARCHAR(100),
+  session_valid BOOLEAN,
+  session_tasks_remaining INT,
+  session_minutes_remaining INT
+) AS $$
+BEGIN
+  RETURN QUERY
+  SELECT
+    wr.current_state,
+    wr.current_city,
+    (wr.geo_session_started_at IS NOT NULL AND wr.geo_session_started_at > NOW() - INTERVAL '60 minutes') as session_valid,
+    GREATEST(0, wr.session_max_tasks - wr.session_task_count) as session_tasks_remaining,
+    GREATEST(0, EXTRACT(EPOCH FROM (wr.geo_session_started_at + INTERVAL '60 minutes' - NOW())) / 60)::INT as session_minutes_remaining
+  FROM worker_registry wr
+  WHERE wr.worker_id = p_worker_id;
+END;
+$$ LANGUAGE plpgsql;
+
+-- View for worker thinness per state
+-- Derives states from dispensaries table - no external states table dependency
+CREATE OR REPLACE VIEW worker_state_capacity AS
+WITH active_states AS (
+  -- Get unique states from dispensaries with valid platform IDs
+  SELECT DISTINCT state as code
+  FROM dispensaries
+  WHERE state IS NOT NULL
+    AND platform_dispensary_id IS NOT NULL
+),
+pending_by_state AS (
+  SELECT d.state, COUNT(*) as count
+  FROM worker_tasks t
+  JOIN dispensaries d ON t.dispensary_id = d.id
+  WHERE t.status = 'pending'
+    AND d.state IS NOT NULL
+  GROUP BY d.state
+),
+workers_by_state AS (
+  SELECT
+    current_state,
+    COUNT(*) as count,
+    SUM(GREATEST(0, session_max_tasks - session_task_count)) as remaining_capacity
+  FROM worker_registry
+  WHERE status IN ('active', 'idle')  -- Include both active and idle workers
+    AND preflight_http_status = 'passed'
+    AND current_state IS NOT NULL
+    AND geo_session_started_at > NOW() - INTERVAL '60 minutes'
+  GROUP BY current_state
+)
+SELECT
+  s.code as state,
+  s.code as state_name, -- Use code as name since we don't have a states lookup table
+  COALESCE(p.count, 0) as pending_tasks,
+  COALESCE(w.count, 0) as workers_on_state,
+  COALESCE(w.remaining_capacity, 0) as remaining_capacity,
+  CASE
+    WHEN COALESCE(w.remaining_capacity, 0) = 0 AND COALESCE(p.count, 0) > 0 THEN 'no_coverage'
+    WHEN COALESCE(w.remaining_capacity, 0) < COALESCE(p.count, 0) THEN 'thin'
+    ELSE 'ok'
+  END as status
+FROM active_states s
+LEFT JOIN pending_by_state p ON p.state = s.code
+LEFT JOIN workers_by_state w ON w.current_state = s.code
+ORDER BY COALESCE(p.count, 0) DESC;

backend/migrations/109_worker_identity_pool.sql

@@ -0,0 +1,354 @@
+-- Migration: 109_worker_identity_pool.sql
+-- Description: Identity pool for diverse IP/fingerprint rotation
+-- Created: 2025-12-14
+--
+-- Workers claim identities (IP + fingerprint) from pool.
+-- Each identity used for 3-5 tasks, then cools down 2-3 hours.
+-- This creates natural browsing patterns - same person doesn't hit 20 stores.
+
+-- ============================================================
+-- IDENTITY POOL TABLE
+-- ============================================================
+CREATE TABLE IF NOT EXISTS worker_identities (
+  id SERIAL PRIMARY KEY,
+
+  -- Evomi session controls the IP
+  session_id VARCHAR(100) UNIQUE NOT NULL,
+
+  -- Detected IP from this session
+  ip_address INET,
+
+  -- Geo targeting
+  state_code VARCHAR(2) NOT NULL,
+  city VARCHAR(100),  -- City-level targeting for diversity
+
+  -- Fingerprint data (UA, timezone, locale, device, etc.)
+  fingerprint JSONB NOT NULL,
+
+  -- Timestamps
+  created_at TIMESTAMPTZ DEFAULT NOW(),
+  last_used_at TIMESTAMPTZ,
+  cooldown_until TIMESTAMPTZ,  -- Can't reuse until this time
+
+  -- Usage stats
+  total_tasks_completed INT DEFAULT 0,
+  total_sessions INT DEFAULT 1,  -- How many times this identity has been used
+
+  -- Current state
+  is_active BOOLEAN DEFAULT FALSE,  -- Currently claimed by a worker
+  active_worker_id VARCHAR(100),    -- Which worker has it
+
+  -- Health tracking
+  consecutive_failures INT DEFAULT 0,
+  is_healthy BOOLEAN DEFAULT TRUE   -- Set false if IP gets blocked
+);
+
+-- Indexes for efficient lookups
+CREATE INDEX IF NOT EXISTS idx_worker_identities_state_city
+  ON worker_identities(state_code, city);
+
+CREATE INDEX IF NOT EXISTS idx_worker_identities_available
+  ON worker_identities(state_code, is_active, cooldown_until)
+  WHERE is_healthy = TRUE;
+
+CREATE INDEX IF NOT EXISTS idx_worker_identities_cooldown
+  ON worker_identities(cooldown_until)
+  WHERE is_healthy = TRUE AND is_active = FALSE;
+
+-- ============================================================
+-- METRO AREA MAPPING
+-- For fallback when exact city not available
+-- ============================================================
+CREATE TABLE IF NOT EXISTS metro_areas (
+  id SERIAL PRIMARY KEY,
+  metro_name VARCHAR(100) NOT NULL,
+  state_code VARCHAR(2) NOT NULL,
+  city VARCHAR(100) NOT NULL,
+  is_primary BOOLEAN DEFAULT FALSE,  -- Primary city of the metro
+  UNIQUE(state_code, city)
+);
+
+-- Phoenix Metro Area
+INSERT INTO metro_areas (metro_name, state_code, city, is_primary) VALUES
+  ('Phoenix Metro', 'AZ', 'Phoenix', TRUE),
+  ('Phoenix Metro', 'AZ', 'Mesa', FALSE),
+  ('Phoenix Metro', 'AZ', 'Glendale', FALSE),
+  ('Phoenix Metro', 'AZ', 'Tempe', FALSE),
+  ('Phoenix Metro', 'AZ', 'Scottsdale', FALSE),
+  ('Phoenix Metro', 'AZ', 'Chandler', FALSE),
+  ('Phoenix Metro', 'AZ', 'Peoria', FALSE),
+  ('Phoenix Metro', 'AZ', 'El Mirage', FALSE),
+  ('Phoenix Metro', 'AZ', 'Tolleson', FALSE),
+  ('Phoenix Metro', 'AZ', 'Sun City', FALSE),
+  ('Phoenix Metro', 'AZ', 'Apache Junction', FALSE),
+  ('Phoenix Metro', 'AZ', 'Cave Creek', FALSE),
+  ('Phoenix Metro', 'AZ', 'Gilbert', FALSE),
+  ('Phoenix Metro', 'AZ', 'Surprise', FALSE),
+  ('Phoenix Metro', 'AZ', 'Avondale', FALSE),
+  ('Phoenix Metro', 'AZ', 'Goodyear', FALSE),
+  ('Phoenix Metro', 'AZ', 'Buckeye', FALSE),
+  ('Phoenix Metro', 'AZ', 'Queen Creek', FALSE)
+ON CONFLICT (state_code, city) DO NOTHING;
+
+-- Tucson Metro Area
+INSERT INTO metro_areas (metro_name, state_code, city, is_primary) VALUES
+  ('Tucson Metro', 'AZ', 'Tucson', TRUE),
+  ('Tucson Metro', 'AZ', 'Oro Valley', FALSE),
+  ('Tucson Metro', 'AZ', 'Marana', FALSE),
+  ('Tucson Metro', 'AZ', 'Sahuarita', FALSE),
+  ('Tucson Metro', 'AZ', 'South Tucson', FALSE)
+ON CONFLICT (state_code, city) DO NOTHING;
+
+-- Flagstaff Area
+INSERT INTO metro_areas (metro_name, state_code, city, is_primary) VALUES
+  ('Flagstaff Area', 'AZ', 'Flagstaff', TRUE),
+  ('Flagstaff Area', 'AZ', 'Sedona', FALSE)
+ON CONFLICT (state_code, city) DO NOTHING;
+
+-- Prescott Area
+INSERT INTO metro_areas (metro_name, state_code, city, is_primary) VALUES
+  ('Prescott Area', 'AZ', 'Prescott', TRUE),
+  ('Prescott Area', 'AZ', 'Prescott Valley', FALSE)
+ON CONFLICT (state_code, city) DO NOTHING;
+
+-- ============================================================
+-- FUNCTION: claim_identity
+-- Claims an available identity for a worker
+-- Tries: exact city -> metro area -> any in state -> create new
+-- ============================================================
+CREATE OR REPLACE FUNCTION claim_identity(
+  p_worker_id VARCHAR(100),
+  p_state_code VARCHAR(2),
+  p_city VARCHAR(100) DEFAULT NULL
+) RETURNS worker_identities AS $$
+DECLARE
+  claimed_identity worker_identities;
+  metro_name_val VARCHAR(100);
+  primary_city VARCHAR(100);
+BEGIN
+  -- 1. Try exact city match (if city provided)
+  IF p_city IS NOT NULL THEN
+    UPDATE worker_identities
+    SET is_active = TRUE,
+        active_worker_id = p_worker_id,
+        last_used_at = NOW()
+    WHERE id = (
+      SELECT id FROM worker_identities
+      WHERE state_code = p_state_code
+        AND city = p_city
+        AND is_active = FALSE
+        AND is_healthy = TRUE
+        AND (cooldown_until IS NULL OR cooldown_until < NOW())
+      ORDER BY last_used_at ASC NULLS FIRST
+      LIMIT 1
+      FOR UPDATE SKIP LOCKED
+    )
+    RETURNING * INTO claimed_identity;
+
+    IF claimed_identity.id IS NOT NULL THEN
+      RETURN claimed_identity;
+    END IF;
+  END IF;
+
+  -- 2. Try metro area fallback
+  IF p_city IS NOT NULL THEN
+    -- Find the metro area for this city
+    SELECT ma.metro_name INTO metro_name_val
+    FROM metro_areas ma
+    WHERE ma.state_code = p_state_code AND ma.city = p_city;
+
+    IF metro_name_val IS NOT NULL THEN
+      -- Get primary city of metro
+      SELECT ma.city INTO primary_city
+      FROM metro_areas ma
+      WHERE ma.metro_name = metro_name_val AND ma.is_primary = TRUE;
+
+      -- Try any city in same metro
+      UPDATE worker_identities wi
+      SET is_active = TRUE,
+          active_worker_id = p_worker_id,
+          last_used_at = NOW()
+      WHERE wi.id = (
+        SELECT wi2.id FROM worker_identities wi2
+        JOIN metro_areas ma ON wi2.city = ma.city AND wi2.state_code = ma.state_code
+        WHERE ma.metro_name = metro_name_val
+          AND wi2.is_active = FALSE
+          AND wi2.is_healthy = TRUE
+          AND (wi2.cooldown_until IS NULL OR wi2.cooldown_until < NOW())
+        ORDER BY wi2.last_used_at ASC NULLS FIRST
+        LIMIT 1
+        FOR UPDATE SKIP LOCKED
+      )
+      RETURNING * INTO claimed_identity;
+
+      IF claimed_identity.id IS NOT NULL THEN
+        RETURN claimed_identity;
+      END IF;
+    END IF;
+  END IF;
+
+  -- 3. Try any identity in state
+  UPDATE worker_identities
+  SET is_active = TRUE,
+      active_worker_id = p_worker_id,
+      last_used_at = NOW()
+  WHERE id = (
+    SELECT id FROM worker_identities
+    WHERE state_code = p_state_code
+      AND is_active = FALSE
+      AND is_healthy = TRUE
+      AND (cooldown_until IS NULL OR cooldown_until < NOW())
+    ORDER BY last_used_at ASC NULLS FIRST
+    LIMIT 1
+    FOR UPDATE SKIP LOCKED
+  )
+  RETURNING * INTO claimed_identity;
+
+  -- Return whatever we got (NULL if nothing available - caller should create new)
+  RETURN claimed_identity;
+END;
+$$ LANGUAGE plpgsql;
+
+-- ============================================================
+-- FUNCTION: release_identity
+-- Releases an identity back to pool with cooldown
+-- ============================================================
+CREATE OR REPLACE FUNCTION release_identity(
+  p_identity_id INT,
+  p_tasks_completed INT DEFAULT 0,
+  p_failed BOOLEAN DEFAULT FALSE
+) RETURNS VOID AS $$
+DECLARE
+  cooldown_hours FLOAT;
+BEGIN
+  -- Random cooldown between 2-3 hours for diversity
+  cooldown_hours := 2 + random();  -- 2.0 to 3.0 hours
+
+  UPDATE worker_identities
+  SET is_active = FALSE,
+      active_worker_id = NULL,
+      total_tasks_completed = total_tasks_completed + p_tasks_completed,
+      total_sessions = total_sessions + 1,
+      cooldown_until = NOW() + (cooldown_hours || ' hours')::INTERVAL,
+      consecutive_failures = CASE WHEN p_failed THEN consecutive_failures + 1 ELSE 0 END,
+      is_healthy = CASE WHEN consecutive_failures >= 3 THEN FALSE ELSE TRUE END
+  WHERE id = p_identity_id;
+END;
+$$ LANGUAGE plpgsql;
+
+-- ============================================================
+-- FUNCTION: get_pending_tasks_by_geo
+-- Gets pending tasks grouped by state/city for identity assignment
+-- ============================================================
+CREATE OR REPLACE FUNCTION get_pending_tasks_by_geo(
+  p_limit INT DEFAULT 10
+) RETURNS TABLE (
+  state_code VARCHAR(2),
+  city VARCHAR(100),
+  pending_count BIGINT,
+  available_identities BIGINT
+) AS $$
+BEGIN
+  RETURN QUERY
+  SELECT
+    d.state as state_code,
+    d.city,
+    COUNT(t.id) as pending_count,
+    (
+      SELECT COUNT(*) FROM worker_identities wi
+      WHERE wi.state_code = d.state
+        AND (wi.city = d.city OR wi.city IS NULL)
+        AND wi.is_active = FALSE
+        AND wi.is_healthy = TRUE
+        AND (wi.cooldown_until IS NULL OR wi.cooldown_until < NOW())
+    ) as available_identities
+  FROM worker_tasks t
+  JOIN dispensaries d ON t.dispensary_id = d.id
+  WHERE t.status = 'pending'
+    AND d.state IS NOT NULL
+  GROUP BY d.state, d.city
+  ORDER BY COUNT(t.id) DESC
+  LIMIT p_limit;
+END;
+$$ LANGUAGE plpgsql;
+
+-- ============================================================
+-- FUNCTION: get_tasks_for_identity
+-- Gets tasks matching an identity's geo (same city or metro)
+-- ============================================================
+CREATE OR REPLACE FUNCTION get_tasks_for_identity(
+  p_state_code VARCHAR(2),
+  p_city VARCHAR(100),
+  p_limit INT DEFAULT 5
+) RETURNS TABLE (
+  task_id INT,
+  dispensary_id INT,
+  dispensary_name VARCHAR(255),
+  dispensary_city VARCHAR(100),
+  role VARCHAR(50)
+) AS $$
+DECLARE
+  metro_name_val VARCHAR(100);
+BEGIN
+  -- Find metro area for this city
+  SELECT ma.metro_name INTO metro_name_val
+  FROM metro_areas ma
+  WHERE ma.state_code = p_state_code AND ma.city = p_city;
+
+  RETURN QUERY
+  SELECT
+    t.id as task_id,
+    d.id as dispensary_id,
+    d.name as dispensary_name,
+    d.city as dispensary_city,
+    t.role
+  FROM worker_tasks t
+  JOIN dispensaries d ON t.dispensary_id = d.id
+  WHERE t.status = 'pending'
+    AND d.state = p_state_code
+    AND (
+      -- Exact city match
+      d.city = p_city
+      -- Or same metro area
+      OR (metro_name_val IS NOT NULL AND d.city IN (
+        SELECT ma.city FROM metro_areas ma WHERE ma.metro_name = metro_name_val
+      ))
+      -- Or any in state if no metro
+      OR (metro_name_val IS NULL)
+    )
+  ORDER BY
+    CASE WHEN d.city = p_city THEN 0 ELSE 1 END,  -- Prefer exact city
+    t.priority DESC,
+    t.created_at ASC
+  LIMIT p_limit;
+END;
+$$ LANGUAGE plpgsql;
+
+-- ============================================================
+-- VIEW: identity_pool_status
+-- Overview of identity pool health and availability
+-- ============================================================
+CREATE OR REPLACE VIEW identity_pool_status AS
+SELECT
+  state_code,
+  city,
+  COUNT(*) as total_identities,
+  COUNT(*) FILTER (WHERE is_active) as active,
+  COUNT(*) FILTER (WHERE NOT is_active AND is_healthy AND (cooldown_until IS NULL OR cooldown_until < NOW())) as available,
+  COUNT(*) FILTER (WHERE NOT is_active AND cooldown_until > NOW()) as cooling_down,
+  COUNT(*) FILTER (WHERE NOT is_healthy) as unhealthy,
+  SUM(total_tasks_completed) as total_tasks,
+  AVG(total_tasks_completed)::INT as avg_tasks_per_identity
+FROM worker_identities
+GROUP BY state_code, city
+ORDER BY state_code, city;
+
+-- ============================================================
+-- Comments
+-- ============================================================
+COMMENT ON TABLE worker_identities IS 'Pool of IP/fingerprint identities for worker rotation';
+COMMENT ON TABLE metro_areas IS 'City groupings for geographic fallback matching';
+COMMENT ON FUNCTION claim_identity IS 'Claim an available identity: exact city -> metro -> state -> NULL (create new)';
+COMMENT ON FUNCTION release_identity IS 'Release identity with 2-3 hour random cooldown';
+COMMENT ON FUNCTION get_pending_tasks_by_geo IS 'Get pending task counts by state/city';
+COMMENT ON FUNCTION get_tasks_for_identity IS 'Get tasks matching identity geo (city or metro area)';

backend/migrations/110_trusted_origins.sql

@@ -0,0 +1,92 @@
+-- Migration: 110_trusted_origins.sql
+-- Description: Trusted origins for API access without token
+-- Created: 2024-12-14
+--
+-- Manages which domains, IPs, and patterns can access the API without a Bearer token.
+-- Used by auth middleware to grant 'internal' role to trusted requests.
+
+-- ============================================================
+-- TRUSTED ORIGINS TABLE
+-- ============================================================
+CREATE TABLE IF NOT EXISTS trusted_origins (
+  id SERIAL PRIMARY KEY,
+
+  -- Origin identification
+  name VARCHAR(100) NOT NULL,              -- Friendly name (e.g., "CannaIQ Production")
+  origin_type VARCHAR(20) NOT NULL,        -- 'domain', 'ip', or 'pattern'
+  origin_value VARCHAR(255) NOT NULL,      -- The actual value to match
+
+  -- Metadata
+  description TEXT,                         -- Optional notes
+  active BOOLEAN DEFAULT TRUE,
+
+  -- Tracking
+  created_at TIMESTAMPTZ DEFAULT NOW(),
+  updated_at TIMESTAMPTZ DEFAULT NOW(),
+  created_by INTEGER REFERENCES users(id),
+
+  -- Constraints
+  CONSTRAINT valid_origin_type CHECK (origin_type IN ('domain', 'ip', 'pattern')),
+  UNIQUE(origin_type, origin_value)
+);
+
+-- Index for active lookups (used by auth middleware)
+CREATE INDEX IF NOT EXISTS idx_trusted_origins_active
+  ON trusted_origins(active) WHERE active = TRUE;
+
+-- Updated at trigger
+CREATE OR REPLACE FUNCTION update_trusted_origins_updated_at()
+RETURNS TRIGGER AS $$
+BEGIN
+  NEW.updated_at = NOW();
+  RETURN NEW;
+END;
+$$ LANGUAGE plpgsql;
+
+DROP TRIGGER IF EXISTS trusted_origins_updated_at ON trusted_origins;
+CREATE TRIGGER trusted_origins_updated_at
+  BEFORE UPDATE ON trusted_origins
+  FOR EACH ROW
+  EXECUTE FUNCTION update_trusted_origins_updated_at();
+
+-- ============================================================
+-- SEED DEFAULT TRUSTED ORIGINS
+-- These match the hardcoded fallbacks in middleware.ts
+-- ============================================================
+
+-- Production domains
+INSERT INTO trusted_origins (name, origin_type, origin_value, description) VALUES
+  ('CannaIQ Production', 'domain', 'https://cannaiq.co', 'Main CannaIQ dashboard'),
+  ('CannaIQ Production (www)', 'domain', 'https://www.cannaiq.co', 'Main CannaIQ dashboard with www'),
+  ('FindADispo Production', 'domain', 'https://findadispo.com', 'Consumer dispensary finder'),
+  ('FindADispo Production (www)', 'domain', 'https://www.findadispo.com', 'Consumer dispensary finder with www'),
+  ('Findagram Production', 'domain', 'https://findagram.co', 'Instagram-style cannabis discovery'),
+  ('Findagram Production (www)', 'domain', 'https://www.findagram.co', 'Instagram-style cannabis discovery with www')
+ON CONFLICT (origin_type, origin_value) DO NOTHING;
+
+-- Wildcard patterns
+INSERT INTO trusted_origins (name, origin_type, origin_value, description) VALUES
+  ('CannaBrands Subdomains', 'pattern', '^https://.*\\.cannabrands\\.app$', 'All *.cannabrands.app subdomains'),
+  ('CannaIQ Subdomains', 'pattern', '^https://.*\\.cannaiq\\.co$', 'All *.cannaiq.co subdomains')
+ON CONFLICT (origin_type, origin_value) DO NOTHING;
+
+-- Local development
+INSERT INTO trusted_origins (name, origin_type, origin_value, description) VALUES
+  ('Local API', 'domain', 'http://localhost:3010', 'Local backend API'),
+  ('Local Admin', 'domain', 'http://localhost:8080', 'Local admin dashboard'),
+  ('Local Vite Dev', 'domain', 'http://localhost:5173', 'Vite dev server')
+ON CONFLICT (origin_type, origin_value) DO NOTHING;
+
+-- Trusted IPs (localhost)
+INSERT INTO trusted_origins (name, origin_type, origin_value, description) VALUES
+  ('Localhost IPv4', 'ip', '127.0.0.1', 'Local machine'),
+  ('Localhost IPv6', 'ip', '::1', 'Local machine IPv6'),
+  ('Localhost IPv6 Mapped', 'ip', '::ffff:127.0.0.1', 'IPv6-mapped IPv4 localhost')
+ON CONFLICT (origin_type, origin_value) DO NOTHING;
+
+-- ============================================================
+-- COMMENTS
+-- ============================================================
+COMMENT ON TABLE trusted_origins IS 'Domains, IPs, and patterns that can access API without token';
+COMMENT ON COLUMN trusted_origins.origin_type IS 'domain = exact URL match, ip = IP address, pattern = regex pattern';
+COMMENT ON COLUMN trusted_origins.origin_value IS 'For domain: full URL. For ip: IP address. For pattern: regex string';

backend/migrations/111_system_settings.sql

@@ -0,0 +1,35 @@
+-- Migration: 111_system_settings.sql
+-- Description: System settings table for runtime configuration
+-- Created: 2024-12-14
+
+CREATE TABLE IF NOT EXISTS system_settings (
+  key VARCHAR(100) PRIMARY KEY,
+  value TEXT NOT NULL,
+  description TEXT,
+  updated_at TIMESTAMPTZ DEFAULT NOW(),
+  updated_by INTEGER REFERENCES users(id)
+);
+
+-- Task pool gate - controls whether workers can claim tasks
+INSERT INTO system_settings (key, value, description) VALUES
+  ('task_pool_open', 'true', 'When false, workers cannot claim new tasks from the pool')
+ON CONFLICT (key) DO NOTHING;
+
+-- Updated at trigger
+CREATE OR REPLACE FUNCTION update_system_settings_updated_at()
+RETURNS TRIGGER AS $$
+BEGIN
+  NEW.updated_at = NOW();
+  RETURN NEW;
+END;
+$$ LANGUAGE plpgsql;
+
+DROP TRIGGER IF EXISTS system_settings_updated_at ON system_settings;
+CREATE TRIGGER system_settings_updated_at
+  BEFORE UPDATE ON system_settings
+  FOR EACH ROW
+  EXECUTE FUNCTION update_system_settings_updated_at();
+
+COMMENT ON TABLE system_settings IS 'Runtime configuration settings';
+COMMENT ON COLUMN system_settings.key IS 'Setting name (e.g., task_pool_open)';
+COMMENT ON COLUMN system_settings.value IS 'Setting value as string';

backend/migrations/112_worker_session_pool.sql

@@ -0,0 +1,390 @@
+-- Migration 112: Worker Session Pool
+-- Tracks IP/fingerprint sessions with exclusive locks and cooldowns
+-- Each worker claims up to 6 tasks, uses one IP/fingerprint for those tasks,
+-- then retires the session (8hr cooldown before IP can be reused)
+
+-- Drop old identity pool tables if they exist (replacing with simpler session model)
+DROP TABLE IF EXISTS worker_identity_claims CASCADE;
+DROP TABLE IF EXISTS worker_identities CASCADE;
+
+-- Worker sessions: tracks active and cooling down IP/fingerprint pairs
+CREATE TABLE IF NOT EXISTS worker_sessions (
+  id SERIAL PRIMARY KEY,
+
+  -- IP and fingerprint for this session
+  ip_address VARCHAR(45) NOT NULL,
+  fingerprint_hash VARCHAR(64) NOT NULL,
+  fingerprint_data JSONB,
+
+  -- Geo this session is locked to
+  state_code VARCHAR(2) NOT NULL,
+  city VARCHAR(100),
+
+  -- Ownership
+  worker_id VARCHAR(255),  -- NULL if in cooldown
+
+  -- Status: 'active' (locked to worker), 'cooldown' (8hr wait), 'available'
+  status VARCHAR(20) NOT NULL DEFAULT 'available',
+
+  -- Task tracking
+  tasks_claimed INTEGER NOT NULL DEFAULT 0,
+  tasks_completed INTEGER NOT NULL DEFAULT 0,
+  tasks_failed INTEGER NOT NULL DEFAULT 0,
+  max_tasks INTEGER NOT NULL DEFAULT 6,
+
+  -- Timestamps
+  created_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
+  locked_at TIMESTAMPTZ,       -- When worker locked this session
+  retired_at TIMESTAMPTZ,      -- When session was retired (cooldown starts)
+  cooldown_until TIMESTAMPTZ,  -- When session becomes available again
+
+  -- Constraints
+  CONSTRAINT valid_status CHECK (status IN ('active', 'cooldown', 'available'))
+);
+
+-- Indexes for fast lookups
+CREATE INDEX IF NOT EXISTS idx_worker_sessions_ip ON worker_sessions(ip_address);
+CREATE INDEX IF NOT EXISTS idx_worker_sessions_status ON worker_sessions(status);
+CREATE INDEX IF NOT EXISTS idx_worker_sessions_worker ON worker_sessions(worker_id) WHERE worker_id IS NOT NULL;
+CREATE INDEX IF NOT EXISTS idx_worker_sessions_geo ON worker_sessions(state_code, city);
+CREATE INDEX IF NOT EXISTS idx_worker_sessions_cooldown ON worker_sessions(cooldown_until) WHERE status = 'cooldown';
+
+-- Unique constraint: only one active session per IP
+CREATE UNIQUE INDEX IF NOT EXISTS idx_worker_sessions_active_ip
+  ON worker_sessions(ip_address)
+  WHERE status = 'active';
+
+-- Function: Check if IP is available (not active, not in cooldown)
+CREATE OR REPLACE FUNCTION is_ip_available(check_ip VARCHAR(45))
+RETURNS BOOLEAN AS $$
+BEGIN
+  -- Check if any session has this IP and is either active or in cooldown
+  RETURN NOT EXISTS (
+    SELECT 1 FROM worker_sessions
+    WHERE ip_address = check_ip
+      AND (status = 'active' OR (status = 'cooldown' AND cooldown_until > NOW()))
+  );
+END;
+$$ LANGUAGE plpgsql;
+
+-- Function: Lock a session to a worker
+-- Returns the session if successful, NULL if IP not available
+CREATE OR REPLACE FUNCTION lock_worker_session(
+  p_worker_id VARCHAR(255),
+  p_ip_address VARCHAR(45),
+  p_state_code VARCHAR(2),
+  p_city VARCHAR(100) DEFAULT NULL,
+  p_fingerprint_hash VARCHAR(64) DEFAULT NULL,
+  p_fingerprint_data JSONB DEFAULT NULL
+) RETURNS worker_sessions AS $$
+DECLARE
+  v_session worker_sessions;
+BEGIN
+  -- First check if IP is available
+  IF NOT is_ip_available(p_ip_address) THEN
+    RETURN NULL;
+  END IF;
+
+  -- Try to find an existing available session for this IP
+  SELECT * INTO v_session
+  FROM worker_sessions
+  WHERE ip_address = p_ip_address
+    AND status = 'available'
+  FOR UPDATE SKIP LOCKED
+  LIMIT 1;
+
+  IF v_session.id IS NOT NULL THEN
+    -- Reuse existing session
+    UPDATE worker_sessions SET
+      worker_id = p_worker_id,
+      status = 'active',
+      state_code = p_state_code,
+      city = p_city,
+      fingerprint_hash = COALESCE(p_fingerprint_hash, fingerprint_hash),
+      fingerprint_data = COALESCE(p_fingerprint_data, fingerprint_data),
+      tasks_claimed = 0,
+      tasks_completed = 0,
+      tasks_failed = 0,
+      locked_at = NOW(),
+      retired_at = NULL,
+      cooldown_until = NULL
+    WHERE id = v_session.id
+    RETURNING * INTO v_session;
+  ELSE
+    -- Create new session
+    INSERT INTO worker_sessions (
+      ip_address, fingerprint_hash, fingerprint_data,
+      state_code, city, worker_id, status, locked_at
+    ) VALUES (
+      p_ip_address, COALESCE(p_fingerprint_hash, md5(random()::text)),
+      p_fingerprint_data, p_state_code, p_city, p_worker_id, 'active', NOW()
+    )
+    RETURNING * INTO v_session;
+  END IF;
+
+  RETURN v_session;
+END;
+$$ LANGUAGE plpgsql;
+
+-- Function: Retire a session (start 8hr cooldown)
+CREATE OR REPLACE FUNCTION retire_worker_session(p_worker_id VARCHAR(255))
+RETURNS BOOLEAN AS $$
+DECLARE
+  v_updated INTEGER;
+BEGIN
+  UPDATE worker_sessions SET
+    status = 'cooldown',
+    worker_id = NULL,
+    retired_at = NOW(),
+    cooldown_until = NOW() + INTERVAL '8 hours'
+  WHERE worker_id = p_worker_id
+    AND status = 'active';
+
+  GET DIAGNOSTICS v_updated = ROW_COUNT;
+  RETURN v_updated > 0;
+END;
+$$ LANGUAGE plpgsql;
+
+-- Function: Release expired cooldowns
+CREATE OR REPLACE FUNCTION release_expired_sessions()
+RETURNS INTEGER AS $$
+DECLARE
+  v_released INTEGER;
+BEGIN
+  UPDATE worker_sessions SET
+    status = 'available'
+  WHERE status = 'cooldown'
+    AND cooldown_until <= NOW();
+
+  GET DIAGNOSTICS v_released = ROW_COUNT;
+  RETURN v_released;
+END;
+$$ LANGUAGE plpgsql;
+
+-- Function: Get session for worker
+CREATE OR REPLACE FUNCTION get_worker_session(p_worker_id VARCHAR(255))
+RETURNS worker_sessions AS $$
+  SELECT * FROM worker_sessions
+  WHERE worker_id = p_worker_id AND status = 'active'
+  LIMIT 1;
+$$ LANGUAGE sql;
+
+-- Function: Increment task counters
+CREATE OR REPLACE FUNCTION session_task_completed(p_worker_id VARCHAR(255))
+RETURNS BOOLEAN AS $$
+BEGIN
+  UPDATE worker_sessions SET
+    tasks_completed = tasks_completed + 1
+  WHERE worker_id = p_worker_id AND status = 'active';
+  RETURN FOUND;
+END;
+$$ LANGUAGE plpgsql;
+
+CREATE OR REPLACE FUNCTION session_task_failed(p_worker_id VARCHAR(255))
+RETURNS BOOLEAN AS $$
+BEGIN
+  UPDATE worker_sessions SET
+    tasks_failed = tasks_failed + 1
+  WHERE worker_id = p_worker_id AND status = 'active';
+  RETURN FOUND;
+END;
+$$ LANGUAGE plpgsql;
+
+CREATE OR REPLACE FUNCTION session_task_claimed(p_worker_id VARCHAR(255), p_count INTEGER DEFAULT 1)
+RETURNS BOOLEAN AS $$
+BEGIN
+  UPDATE worker_sessions SET
+    tasks_claimed = tasks_claimed + p_count
+  WHERE worker_id = p_worker_id AND status = 'active';
+  RETURN FOUND;
+END;
+$$ LANGUAGE plpgsql;
+
+-- Scheduled job hint: Run release_expired_sessions() every 5 minutes
+COMMENT ON FUNCTION release_expired_sessions() IS
+  'Run periodically to release sessions from cooldown. Suggest: every 5 minutes.';
+
+-- =============================================================================
+-- ATOMIC TASK CLAIMING
+-- Worker claims up to 6 tasks for same geo in one transaction
+-- =============================================================================
+
+-- Function: Claim up to N tasks for same geo
+-- Returns claimed tasks with dispensary geo info
+CREATE OR REPLACE FUNCTION claim_tasks_batch(
+  p_worker_id VARCHAR(255),
+  p_max_tasks INTEGER DEFAULT 6,
+  p_role VARCHAR(50) DEFAULT NULL  -- Optional role filter
+) RETURNS TABLE (
+  task_id INTEGER,
+  role VARCHAR(50),
+  dispensary_id INTEGER,
+  dispensary_name VARCHAR(255),
+  city VARCHAR(100),
+  state_code VARCHAR(2),
+  platform VARCHAR(50),
+  method VARCHAR(20)
+) AS $$
+DECLARE
+  v_target_state VARCHAR(2);
+  v_target_city VARCHAR(100);
+  v_claimed_count INTEGER := 0;
+BEGIN
+  -- First, find the geo with most pending tasks to target
+  SELECT d.state, d.city INTO v_target_state, v_target_city
+  FROM worker_tasks t
+  JOIN dispensaries d ON t.dispensary_id = d.id
+  WHERE t.status = 'pending'
+    AND (p_role IS NULL OR t.role = p_role)
+  GROUP BY d.state, d.city
+  ORDER BY COUNT(*) DESC
+  LIMIT 1;
+
+  -- No pending tasks
+  IF v_target_state IS NULL THEN
+    RETURN;
+  END IF;
+
+  -- Claim up to p_max_tasks for this geo
+  RETURN QUERY
+  WITH claimed AS (
+    UPDATE worker_tasks t SET
+      status = 'claimed',
+      worker_id = p_worker_id,
+      claimed_at = NOW()
+    FROM (
+      SELECT t2.id
+      FROM worker_tasks t2
+      JOIN dispensaries d ON t2.dispensary_id = d.id
+      WHERE t2.status = 'pending'
+        AND d.state = v_target_state
+        AND (v_target_city IS NULL OR d.city = v_target_city)
+        AND (p_role IS NULL OR t2.role = p_role)
+      ORDER BY t2.priority DESC, t2.created_at ASC
+      FOR UPDATE SKIP LOCKED
+      LIMIT p_max_tasks
+    ) sub
+    WHERE t.id = sub.id
+    RETURNING t.id, t.role, t.dispensary_id, t.method
+  )
+  SELECT
+    c.id as task_id,
+    c.role,
+    c.dispensary_id,
+    d.name as dispensary_name,
+    d.city,
+    d.state as state_code,
+    d.platform,
+    c.method
+  FROM claimed c
+  JOIN dispensaries d ON c.dispensary_id = d.id;
+END;
+$$ LANGUAGE plpgsql;
+
+-- Function: Release claimed tasks back to pending (for failed worker or cleanup)
+CREATE OR REPLACE FUNCTION release_claimed_tasks(p_worker_id VARCHAR(255))
+RETURNS INTEGER AS $$
+DECLARE
+  v_released INTEGER;
+BEGIN
+  UPDATE worker_tasks SET
+    status = 'pending',
+    worker_id = NULL,
+    claimed_at = NULL
+  WHERE worker_id = p_worker_id
+    AND status IN ('claimed', 'running');
+
+  GET DIAGNOSTICS v_released = ROW_COUNT;
+  RETURN v_released;
+END;
+$$ LANGUAGE plpgsql;
+
+-- Function: Mark task as running
+CREATE OR REPLACE FUNCTION start_task(p_task_id INTEGER, p_worker_id VARCHAR(255))
+RETURNS BOOLEAN AS $$
+BEGIN
+  UPDATE worker_tasks SET
+    status = 'running',
+    started_at = NOW()
+  WHERE id = p_task_id
+    AND worker_id = p_worker_id
+    AND status = 'claimed';
+  RETURN FOUND;
+END;
+$$ LANGUAGE plpgsql;
+
+-- Function: Mark task as completed (leaves pool)
+CREATE OR REPLACE FUNCTION complete_task(
+  p_task_id INTEGER,
+  p_worker_id VARCHAR(255),
+  p_result JSONB DEFAULT NULL
+) RETURNS BOOLEAN AS $$
+BEGIN
+  UPDATE worker_tasks SET
+    status = 'completed',
+    completed_at = NOW(),
+    result = p_result
+  WHERE id = p_task_id
+    AND worker_id = p_worker_id
+    AND status = 'running';
+  RETURN FOUND;
+END;
+$$ LANGUAGE plpgsql;
+
+-- Function: Mark task as failed (returns to pending for retry)
+CREATE OR REPLACE FUNCTION fail_task(
+  p_task_id INTEGER,
+  p_worker_id VARCHAR(255),
+  p_error TEXT DEFAULT NULL,
+  p_max_retries INTEGER DEFAULT 3
+) RETURNS BOOLEAN AS $$
+DECLARE
+  v_retry_count INTEGER;
+BEGIN
+  -- Get current retry count
+  SELECT COALESCE(retry_count, 0) INTO v_retry_count
+  FROM worker_tasks WHERE id = p_task_id;
+
+  IF v_retry_count >= p_max_retries THEN
+    -- Max retries exceeded - mark as permanently failed
+    UPDATE worker_tasks SET
+      status = 'failed',
+      completed_at = NOW(),
+      error_message = p_error,
+      retry_count = v_retry_count + 1
+    WHERE id = p_task_id
+      AND worker_id = p_worker_id;
+  ELSE
+    -- Return to pending for retry
+    UPDATE worker_tasks SET
+      status = 'pending',
+      worker_id = NULL,
+      claimed_at = NULL,
+      started_at = NULL,
+      error_message = p_error,
+      retry_count = v_retry_count + 1
+    WHERE id = p_task_id
+      AND worker_id = p_worker_id;
+  END IF;
+
+  RETURN FOUND;
+END;
+$$ LANGUAGE plpgsql;
+
+-- Add retry_count column if not exists
+DO $$
+BEGIN
+  IF NOT EXISTS (
+    SELECT 1 FROM information_schema.columns
+    WHERE table_name = 'worker_tasks' AND column_name = 'retry_count'
+  ) THEN
+    ALTER TABLE worker_tasks ADD COLUMN retry_count INTEGER NOT NULL DEFAULT 0;
+  END IF;
+
+  IF NOT EXISTS (
+    SELECT 1 FROM information_schema.columns
+    WHERE table_name = 'worker_tasks' AND column_name = 'claimed_at'
+  ) THEN
+    ALTER TABLE worker_tasks ADD COLUMN claimed_at TIMESTAMPTZ;
+  END IF;
+END $$;

backend/migrations/113_task_pools.sql

@@ -0,0 +1,381 @@
+-- Task Pools: Group tasks by geo area for worker assignment
+-- Workers claim a pool, get proxy for that geo, then pull tasks from pool
+
+-- ============================================================================
+-- TASK POOLS TABLE
+-- ============================================================================
+-- Each pool represents a metro area (e.g., Phoenix AZ = 100mi radius)
+-- Dispensaries are assigned to pools based on location
+-- Workers claim a pool, not individual tasks
+
+CREATE TABLE IF NOT EXISTS task_pools (
+  id SERIAL PRIMARY KEY,
+  name VARCHAR(100) NOT NULL UNIQUE,        -- e.g., 'phoenix_az'
+  display_name VARCHAR(100) NOT NULL,       -- e.g., 'Phoenix, AZ'
+  state_code VARCHAR(2) NOT NULL,           -- e.g., 'AZ'
+  city VARCHAR(100) NOT NULL,               -- e.g., 'Phoenix'
+  latitude DECIMAL(10, 6) NOT NULL,         -- pool center lat
+  longitude DECIMAL(10, 6) NOT NULL,        -- pool center lng
+  radius_miles INTEGER DEFAULT 100,         -- pool radius (100mi default)
+  timezone VARCHAR(50) NOT NULL,            -- e.g., 'America/Phoenix'
+  is_active BOOLEAN DEFAULT true,
+  created_at TIMESTAMPTZ DEFAULT NOW(),
+  updated_at TIMESTAMPTZ DEFAULT NOW()
+);
+
+-- Index for active pools
+CREATE INDEX IF NOT EXISTS idx_task_pools_active ON task_pools(is_active) WHERE is_active = true;
+
+-- ============================================================================
+-- LINK DISPENSARIES TO POOLS
+-- ============================================================================
+-- Add pool_id to dispensaries table
+
+ALTER TABLE dispensaries
+ADD COLUMN IF NOT EXISTS pool_id INTEGER REFERENCES task_pools(id);
+
+-- Index for pool membership
+CREATE INDEX IF NOT EXISTS idx_dispensaries_pool ON dispensaries(pool_id) WHERE pool_id IS NOT NULL;
+
+-- ============================================================================
+-- WORKER POOL ASSIGNMENT
+-- ============================================================================
+-- Track which pool a worker is currently assigned to
+
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS current_pool_id INTEGER REFERENCES task_pools(id),
+ADD COLUMN IF NOT EXISTS pool_claimed_at TIMESTAMPTZ,
+ADD COLUMN IF NOT EXISTS pool_stores_visited INTEGER DEFAULT 0,
+ADD COLUMN IF NOT EXISTS pool_max_stores INTEGER DEFAULT 6;
+
+-- ============================================================================
+-- SEED INITIAL POOLS
+-- ============================================================================
+-- Major cannabis markets with approximate center coordinates
+
+INSERT INTO task_pools (name, display_name, state_code, city, latitude, longitude, timezone, radius_miles) VALUES
+  -- Arizona
+  ('phoenix_az', 'Phoenix, AZ', 'AZ', 'Phoenix', 33.4484, -112.0740, 'America/Phoenix', 100),
+  ('tucson_az', 'Tucson, AZ', 'AZ', 'Tucson', 32.2226, -110.9747, 'America/Phoenix', 75),
+
+  -- California
+  ('los_angeles_ca', 'Los Angeles, CA', 'CA', 'Los Angeles', 34.0522, -118.2437, 'America/Los_Angeles', 100),
+  ('san_francisco_ca', 'San Francisco, CA', 'CA', 'San Francisco', 37.7749, -122.4194, 'America/Los_Angeles', 75),
+  ('san_diego_ca', 'San Diego, CA', 'CA', 'San Diego', 32.7157, -117.1611, 'America/Los_Angeles', 75),
+  ('sacramento_ca', 'Sacramento, CA', 'CA', 'Sacramento', 38.5816, -121.4944, 'America/Los_Angeles', 75),
+
+  -- Colorado
+  ('denver_co', 'Denver, CO', 'CO', 'Denver', 39.7392, -104.9903, 'America/Denver', 100),
+
+  -- Illinois
+  ('chicago_il', 'Chicago, IL', 'IL', 'Chicago', 41.8781, -87.6298, 'America/Chicago', 100),
+
+  -- Massachusetts
+  ('boston_ma', 'Boston, MA', 'MA', 'Boston', 42.3601, -71.0589, 'America/New_York', 75),
+
+  -- Michigan
+  ('detroit_mi', 'Detroit, MI', 'MI', 'Detroit', 42.3314, -83.0458, 'America/Detroit', 100),
+
+  -- Nevada
+  ('las_vegas_nv', 'Las Vegas, NV', 'NV', 'Las Vegas', 36.1699, -115.1398, 'America/Los_Angeles', 75),
+  ('reno_nv', 'Reno, NV', 'NV', 'Reno', 39.5296, -119.8138, 'America/Los_Angeles', 50),
+
+  -- New Jersey
+  ('newark_nj', 'Newark, NJ', 'NJ', 'Newark', 40.7357, -74.1724, 'America/New_York', 75),
+
+  -- New York
+  ('new_york_ny', 'New York, NY', 'NY', 'New York', 40.7128, -74.0060, 'America/New_York', 75),
+
+  -- Oklahoma
+  ('oklahoma_city_ok', 'Oklahoma City, OK', 'OK', 'Oklahoma City', 35.4676, -97.5164, 'America/Chicago', 100),
+  ('tulsa_ok', 'Tulsa, OK', 'OK', 'Tulsa', 36.1540, -95.9928, 'America/Chicago', 75),
+
+  -- Oregon
+  ('portland_or', 'Portland, OR', 'OR', 'Portland', 45.5152, -122.6784, 'America/Los_Angeles', 75),
+
+  -- Washington
+  ('seattle_wa', 'Seattle, WA', 'WA', 'Seattle', 47.6062, -122.3321, 'America/Los_Angeles', 100)
+
+ON CONFLICT (name) DO NOTHING;
+
+-- ============================================================================
+-- FUNCTION: Assign dispensary to nearest pool
+-- ============================================================================
+CREATE OR REPLACE FUNCTION assign_dispensary_to_pool(disp_id INTEGER)
+RETURNS INTEGER AS $$
+DECLARE
+  disp_lat DECIMAL(10,6);
+  disp_lng DECIMAL(10,6);
+  nearest_pool_id INTEGER;
+BEGIN
+  -- Get dispensary coordinates
+  SELECT latitude, longitude INTO disp_lat, disp_lng
+  FROM dispensaries WHERE id = disp_id;
+
+  IF disp_lat IS NULL OR disp_lng IS NULL THEN
+    RETURN NULL;
+  END IF;
+
+  -- Find nearest active pool within radius
+  -- Using Haversine approximation (accurate enough for 100mi)
+  SELECT id INTO nearest_pool_id
+  FROM task_pools
+  WHERE is_active = true
+    AND (
+      3959 * acos(
+        cos(radians(latitude)) * cos(radians(disp_lat)) *
+        cos(radians(disp_lng) - radians(longitude)) +
+        sin(radians(latitude)) * sin(radians(disp_lat))
+      )
+    ) <= radius_miles
+  ORDER BY (
+    3959 * acos(
+      cos(radians(latitude)) * cos(radians(disp_lat)) *
+      cos(radians(disp_lng) - radians(longitude)) +
+      sin(radians(latitude)) * sin(radians(disp_lat))
+    )
+  )
+  LIMIT 1;
+
+  -- Update dispensary
+  IF nearest_pool_id IS NOT NULL THEN
+    UPDATE dispensaries SET pool_id = nearest_pool_id WHERE id = disp_id;
+  END IF;
+
+  RETURN nearest_pool_id;
+END;
+$$ LANGUAGE plpgsql;
+
+-- ============================================================================
+-- FUNCTION: Assign all dispensaries to pools (batch)
+-- ============================================================================
+CREATE OR REPLACE FUNCTION assign_all_dispensaries_to_pools()
+RETURNS TABLE(assigned INTEGER, unassigned INTEGER) AS $$
+DECLARE
+  assigned_count INTEGER := 0;
+  unassigned_count INTEGER := 0;
+  disp RECORD;
+  pool_id INTEGER;
+BEGIN
+  FOR disp IN SELECT id FROM dispensaries WHERE pool_id IS NULL AND latitude IS NOT NULL LOOP
+    pool_id := assign_dispensary_to_pool(disp.id);
+    IF pool_id IS NOT NULL THEN
+      assigned_count := assigned_count + 1;
+    ELSE
+      unassigned_count := unassigned_count + 1;
+    END IF;
+  END LOOP;
+
+  RETURN QUERY SELECT assigned_count, unassigned_count;
+END;
+$$ LANGUAGE plpgsql;
+
+-- ============================================================================
+-- FUNCTION: Get pools with pending tasks
+-- ============================================================================
+CREATE OR REPLACE FUNCTION get_pools_with_pending_tasks()
+RETURNS TABLE(
+  pool_id INTEGER,
+  pool_name VARCHAR(100),
+  display_name VARCHAR(100),
+  state_code VARCHAR(2),
+  city VARCHAR(100),
+  timezone VARCHAR(50),
+  pending_count BIGINT,
+  store_count BIGINT
+) AS $$
+BEGIN
+  RETURN QUERY
+  SELECT
+    tp.id as pool_id,
+    tp.name as pool_name,
+    tp.display_name,
+    tp.state_code,
+    tp.city,
+    tp.timezone,
+    COUNT(DISTINCT t.id) as pending_count,
+    COUNT(DISTINCT d.id) as store_count
+  FROM task_pools tp
+  JOIN dispensaries d ON d.pool_id = tp.id
+  JOIN tasks t ON t.dispensary_id = d.id AND t.status = 'pending'
+  WHERE tp.is_active = true
+  GROUP BY tp.id, tp.name, tp.display_name, tp.state_code, tp.city, tp.timezone
+  HAVING COUNT(DISTINCT t.id) > 0
+  ORDER BY COUNT(DISTINCT t.id) DESC;
+END;
+$$ LANGUAGE plpgsql;
+
+-- ============================================================================
+-- FUNCTION: Worker claims a pool
+-- ============================================================================
+CREATE OR REPLACE FUNCTION worker_claim_pool(
+  p_worker_id VARCHAR(100),
+  p_pool_id INTEGER DEFAULT NULL
+)
+RETURNS TABLE(
+  pool_id INTEGER,
+  pool_name VARCHAR(100),
+  display_name VARCHAR(100),
+  state_code VARCHAR(2),
+  city VARCHAR(100),
+  latitude DECIMAL(10,6),
+  longitude DECIMAL(10,6),
+  timezone VARCHAR(50)
+) AS $$
+DECLARE
+  claimed_pool_id INTEGER;
+BEGIN
+  -- If no pool specified, pick the one with most pending tasks
+  IF p_pool_id IS NULL THEN
+    SELECT tp.id INTO claimed_pool_id
+    FROM task_pools tp
+    JOIN dispensaries d ON d.pool_id = tp.id
+    JOIN tasks t ON t.dispensary_id = d.id AND t.status = 'pending'
+    WHERE tp.is_active = true
+    GROUP BY tp.id
+    ORDER BY COUNT(DISTINCT t.id) DESC
+    LIMIT 1;
+  ELSE
+    claimed_pool_id := p_pool_id;
+  END IF;
+
+  IF claimed_pool_id IS NULL THEN
+    RETURN;
+  END IF;
+
+  -- Update worker registry with pool assignment
+  UPDATE worker_registry
+  SET
+    current_pool_id = claimed_pool_id,
+    pool_claimed_at = NOW(),
+    pool_stores_visited = 0,
+    pool_max_stores = 6,
+    updated_at = NOW()
+  WHERE worker_id = p_worker_id;
+
+  -- Return pool info
+  RETURN QUERY
+  SELECT
+    tp.id,
+    tp.name,
+    tp.display_name,
+    tp.state_code,
+    tp.city,
+    tp.latitude,
+    tp.longitude,
+    tp.timezone
+  FROM task_pools tp
+  WHERE tp.id = claimed_pool_id;
+END;
+$$ LANGUAGE plpgsql;
+
+-- ============================================================================
+-- FUNCTION: Pull tasks from worker's pool (up to 6 stores)
+-- ============================================================================
+CREATE OR REPLACE FUNCTION pull_tasks_from_pool(
+  p_worker_id VARCHAR(100),
+  p_max_stores INTEGER DEFAULT 6
+)
+RETURNS TABLE(
+  task_id INTEGER,
+  dispensary_id INTEGER,
+  dispensary_name VARCHAR(255),
+  role VARCHAR(50),
+  platform VARCHAR(50),
+  method VARCHAR(20)
+) AS $$
+DECLARE
+  worker_pool_id INTEGER;
+  stores_visited INTEGER;
+  max_stores INTEGER;
+  stores_remaining INTEGER;
+BEGIN
+  -- Get worker's current pool and store count
+  SELECT current_pool_id, pool_stores_visited, pool_max_stores
+  INTO worker_pool_id, stores_visited, max_stores
+  FROM worker_registry
+  WHERE worker_id = p_worker_id;
+
+  IF worker_pool_id IS NULL THEN
+    RAISE EXCEPTION 'Worker % has no pool assigned', p_worker_id;
+  END IF;
+
+  stores_remaining := max_stores - stores_visited;
+  IF stores_remaining <= 0 THEN
+    RETURN; -- Worker exhausted
+  END IF;
+
+  -- Claim tasks from pool (one task per store, up to remaining capacity)
+  RETURN QUERY
+  WITH available_stores AS (
+    SELECT DISTINCT ON (d.id)
+      t.id as task_id,
+      d.id as dispensary_id,
+      d.name as dispensary_name,
+      t.role,
+      t.platform,
+      t.method
+    FROM tasks t
+    JOIN dispensaries d ON d.id = t.dispensary_id
+    WHERE d.pool_id = worker_pool_id
+      AND t.status = 'pending'
+      AND t.scheduled_for <= NOW()
+    ORDER BY d.id, t.priority DESC, t.created_at ASC
+    LIMIT stores_remaining
+  ),
+  claimed AS (
+    UPDATE tasks
+    SET
+      status = 'claimed',
+      claimed_by = p_worker_id,
+      claimed_at = NOW()
+    WHERE id IN (SELECT task_id FROM available_stores)
+    RETURNING id
+  )
+  SELECT
+    av.task_id,
+    av.dispensary_id,
+    av.dispensary_name,
+    av.role,
+    av.platform,
+    av.method
+  FROM available_stores av
+  WHERE av.task_id IN (SELECT id FROM claimed);
+
+  -- Update worker store count
+  UPDATE worker_registry
+  SET
+    pool_stores_visited = pool_stores_visited + (
+      SELECT COUNT(DISTINCT dispensary_id)
+      FROM tasks
+      WHERE claimed_by = p_worker_id AND status = 'claimed'
+    ),
+    updated_at = NOW()
+  WHERE worker_id = p_worker_id;
+END;
+$$ LANGUAGE plpgsql;
+
+-- ============================================================================
+-- FUNCTION: Worker releases pool (exhausted or done)
+-- ============================================================================
+CREATE OR REPLACE FUNCTION worker_release_pool(p_worker_id VARCHAR(100))
+RETURNS BOOLEAN AS $$
+BEGIN
+  UPDATE worker_registry
+  SET
+    current_pool_id = NULL,
+    pool_claimed_at = NULL,
+    pool_stores_visited = 0,
+    current_state = NULL,
+    current_city = NULL,
+    updated_at = NOW()
+  WHERE worker_id = p_worker_id;
+
+  RETURN true;
+END;
+$$ LANGUAGE plpgsql;
+
+-- ============================================================================
+-- RUN: Assign existing dispensaries to pools
+-- ============================================================================
+SELECT * FROM assign_all_dispensaries_to_pools();

backend/migrations/114_schedule_pool_id.sql

@@ -0,0 +1,10 @@
+-- Migration 114: Add pool_id to task_schedules
+-- Allows schedules to target specific geo pools
+
+ALTER TABLE task_schedules
+ADD COLUMN IF NOT EXISTS pool_id INTEGER REFERENCES task_pools(id);
+
+-- Index for pool-based schedule queries
+CREATE INDEX IF NOT EXISTS idx_task_schedules_pool ON task_schedules(pool_id) WHERE pool_id IS NOT NULL;
+
+COMMENT ON COLUMN task_schedules.pool_id IS 'Optional geo pool filter. NULL = all pools/dispensaries matching state_code';

backend/migrations/115_task_proxy_ip.sql

@@ -0,0 +1,17 @@
+-- Migration: Add proxy_ip tracking to worker_tasks
+-- Purpose: Prevent same IP from hitting multiple stores on same platform simultaneously
+--
+-- Anti-detection measure: Dutchie/Jane may flag if same IP makes requests
+-- for multiple different stores. This column lets us track and prevent that.
+
+-- Add proxy_ip column to track which proxy IP is being used for each task
+ALTER TABLE worker_tasks ADD COLUMN IF NOT EXISTS proxy_ip VARCHAR(45);
+
+-- Index for quick lookup of active tasks by proxy IP
+-- Used to check: "Is this IP already hitting another store?"
+CREATE INDEX IF NOT EXISTS idx_worker_tasks_proxy_ip_active
+ON worker_tasks (proxy_ip, platform)
+WHERE status IN ('claimed', 'running') AND proxy_ip IS NOT NULL;
+
+-- Comment
+COMMENT ON COLUMN worker_tasks.proxy_ip IS 'Proxy IP assigned to this task. Used to prevent same IP hitting multiple stores on same platform.';

backend/migrations/116_task_source_tracking.sql

@@ -0,0 +1,16 @@
+-- Migration: Add source tracking columns to worker_tasks
+-- Purpose: Track where tasks originated from (schedule, API, manual)
+
+-- Add source tracking columns
+ALTER TABLE worker_tasks ADD COLUMN IF NOT EXISTS source VARCHAR(50);
+ALTER TABLE worker_tasks ADD COLUMN IF NOT EXISTS source_schedule_id INTEGER REFERENCES task_schedules(id);
+ALTER TABLE worker_tasks ADD COLUMN IF NOT EXISTS source_metadata JSONB;
+
+-- Index for tracking tasks by schedule
+CREATE INDEX IF NOT EXISTS idx_worker_tasks_source_schedule
+ON worker_tasks (source_schedule_id) WHERE source_schedule_id IS NOT NULL;
+
+-- Comments
+COMMENT ON COLUMN worker_tasks.source IS 'Origin of task: schedule, api, manual, chain';
+COMMENT ON COLUMN worker_tasks.source_schedule_id IS 'ID of schedule that created this task';
+COMMENT ON COLUMN worker_tasks.source_metadata IS 'Additional metadata about task origin';

backend/migrations/117_per_store_crawl_interval.sql

@@ -0,0 +1,32 @@
+-- Migration 117: Per-store crawl interval scheduling
+-- Adds columns for configurable per-store crawl intervals
+-- Part of Real-Time Inventory Tracking feature
+
+-- Per-store crawl interval (NULL = use state schedule default 4h)
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS crawl_interval_minutes INT DEFAULT NULL;
+
+-- When this store should next be crawled (used by high-frequency scheduler)
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS next_crawl_at TIMESTAMPTZ DEFAULT NULL;
+
+-- Track last request time to enforce minimum spacing
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS last_crawl_started_at TIMESTAMPTZ DEFAULT NULL;
+
+-- Change tracking for optimization
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS last_inventory_hash TEXT DEFAULT NULL;
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS last_price_hash TEXT DEFAULT NULL;
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS inventory_changes_24h INT DEFAULT 0;
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS price_changes_24h INT DEFAULT 0;
+
+-- Index for scheduler query: find stores due for high-frequency crawl
+CREATE INDEX IF NOT EXISTS idx_dispensaries_next_crawl
+ON dispensaries(next_crawl_at)
+WHERE crawl_interval_minutes IS NOT NULL AND crawl_enabled = TRUE;
+
+-- Comment for documentation
+COMMENT ON COLUMN dispensaries.crawl_interval_minutes IS 'Custom crawl interval in minutes. NULL = use state schedule (4h default). Set to 15/30/60 for high-frequency tracking.';
+COMMENT ON COLUMN dispensaries.next_crawl_at IS 'When this store should next be crawled. Updated after each crawl with interval + jitter.';
+COMMENT ON COLUMN dispensaries.last_crawl_started_at IS 'When the last crawl task was created. Used to enforce minimum spacing.';
+COMMENT ON COLUMN dispensaries.last_inventory_hash IS 'Hash of inventory state from last crawl. Used to detect changes and skip unchanged payloads.';
+COMMENT ON COLUMN dispensaries.last_price_hash IS 'Hash of price state from last crawl. Used to detect price changes.';
+COMMENT ON COLUMN dispensaries.inventory_changes_24h IS 'Number of inventory changes detected in last 24h. Indicates store volatility.';
+COMMENT ON COLUMN dispensaries.price_changes_24h IS 'Number of price changes detected in last 24h.';

backend/migrations/118_inventory_snapshots.sql

@@ -0,0 +1,48 @@
+-- Migration 118: Inventory snapshots table
+-- Lightweight per-product tracking for sales velocity estimation
+-- Part of Real-Time Inventory Tracking feature
+
+CREATE TABLE IF NOT EXISTS inventory_snapshots (
+  id BIGSERIAL PRIMARY KEY,
+  dispensary_id INT NOT NULL REFERENCES dispensaries(id) ON DELETE CASCADE,
+  product_id TEXT NOT NULL,           -- provider_product_id (normalized across platforms)
+  captured_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
+
+  -- Platform (for debugging/filtering)
+  platform TEXT NOT NULL,             -- 'dutchie' | 'jane' | 'treez'
+
+  -- Inventory fields (normalized from all platforms)
+  quantity_available INT,             -- Dutchie: quantityAvailable, Jane: quantity, Treez: quantityAvailable
+  is_below_threshold BOOLEAN,         -- Dutchie: isBelowThreshold, Jane: computed, Treez: lowInventory
+  status TEXT,                        -- Active/Inactive/available
+
+  -- Price fields (normalized)
+  price_rec NUMERIC(10,2),            -- recreational price
+  price_med NUMERIC(10,2),            -- medical price (if different)
+
+  -- Denormalized for fast queries
+  brand_name TEXT,
+  category TEXT,
+  product_name TEXT
+);
+
+-- Primary query: get snapshots for a store over time
+CREATE INDEX idx_inv_snap_store_time ON inventory_snapshots(dispensary_id, captured_at DESC);
+
+-- Delta calculation: get consecutive snapshots for a product
+CREATE INDEX idx_inv_snap_product_time ON inventory_snapshots(dispensary_id, product_id, captured_at DESC);
+
+-- Brand-level analytics
+CREATE INDEX idx_inv_snap_brand_time ON inventory_snapshots(brand_name, captured_at DESC) WHERE brand_name IS NOT NULL;
+
+-- Platform filtering
+CREATE INDEX idx_inv_snap_platform ON inventory_snapshots(platform, captured_at DESC);
+
+-- Retention cleanup (30 days) - simple index, cleanup job handles the WHERE
+CREATE INDEX IF NOT EXISTS idx_inv_snap_cleanup ON inventory_snapshots(captured_at);
+
+-- Comments
+COMMENT ON TABLE inventory_snapshots IS 'Lightweight inventory snapshots for sales velocity tracking. Retained 30 days.';
+COMMENT ON COLUMN inventory_snapshots.product_id IS 'Provider product ID, normalized across platforms';
+COMMENT ON COLUMN inventory_snapshots.platform IS 'Menu platform: dutchie, jane, or treez';
+COMMENT ON COLUMN inventory_snapshots.quantity_available IS 'Current quantity in stock (Dutchie: quantityAvailable, Jane: quantity)';

backend/migrations/119_product_visibility_events.sql

@@ -0,0 +1,53 @@
+-- Migration 119: Product visibility events table
+-- Tracks OOS, brand drops, and other notable events for alerts
+-- Part of Real-Time Inventory Tracking feature
+
+CREATE TABLE IF NOT EXISTS product_visibility_events (
+  id SERIAL PRIMARY KEY,
+  dispensary_id INT NOT NULL REFERENCES dispensaries(id) ON DELETE CASCADE,
+
+  -- Product identification (null for brand-level events)
+  product_id TEXT,                    -- provider_product_id
+  product_name TEXT,                  -- For display in alerts
+
+  -- Brand (always populated)
+  brand_name TEXT,
+
+  -- Event details
+  event_type TEXT NOT NULL,           -- 'oos', 'back_in_stock', 'brand_dropped', 'brand_added', 'price_change'
+  detected_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
+
+  -- Context
+  previous_quantity INT,              -- For OOS events: what quantity was before
+  previous_price NUMERIC(10,2),       -- For price change events
+  new_price NUMERIC(10,2),            -- For price change events
+  price_change_pct NUMERIC(5,2),      -- Percentage change (e.g., -15.5 for 15.5% decrease)
+
+  -- Platform
+  platform TEXT,                      -- 'dutchie' | 'jane' | 'treez'
+
+  -- Alert status
+  notified BOOLEAN DEFAULT FALSE,     -- Has external system been notified?
+  acknowledged_at TIMESTAMPTZ,        -- When user acknowledged the alert
+  acknowledged_by TEXT                -- User who acknowledged
+);
+
+-- Primary query: recent events by store
+CREATE INDEX idx_vis_events_store_time ON product_visibility_events(dispensary_id, detected_at DESC);
+
+-- Alert queries: unnotified events
+CREATE INDEX idx_vis_events_unnotified ON product_visibility_events(notified, detected_at DESC) WHERE notified = FALSE;
+
+-- Event type filtering
+CREATE INDEX idx_vis_events_type ON product_visibility_events(event_type, detected_at DESC);
+
+-- Brand-level queries
+CREATE INDEX idx_vis_events_brand ON product_visibility_events(brand_name, event_type, detected_at DESC) WHERE brand_name IS NOT NULL;
+
+-- Cleanup (90 days retention) - simple index, cleanup job handles the WHERE
+CREATE INDEX IF NOT EXISTS idx_vis_events_cleanup ON product_visibility_events(detected_at);
+
+-- Comments
+COMMENT ON TABLE product_visibility_events IS 'Notable inventory events for alerting. OOS, brand drops, significant price changes. Retained 90 days.';
+COMMENT ON COLUMN product_visibility_events.event_type IS 'Event type: oos (out of stock), back_in_stock, brand_dropped, brand_added, price_change';
+COMMENT ON COLUMN product_visibility_events.notified IS 'Whether external systems (other apps) have been notified of this event';

backend/migrations/120_daily_baseline_tracking.sql

@@ -0,0 +1,13 @@
+-- Migration 120: Daily baseline tracking
+-- Track when each store's daily baseline payload was last saved
+-- Part of Real-Time Inventory Tracking feature
+
+-- Add column to track last baseline save time
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS last_baseline_at TIMESTAMPTZ DEFAULT NULL;
+
+-- Index for finding stores that need baselines
+CREATE INDEX IF NOT EXISTS idx_dispensaries_baseline ON dispensaries(last_baseline_at)
+WHERE crawl_enabled = TRUE;
+
+-- Comment
+COMMENT ON COLUMN dispensaries.last_baseline_at IS 'Timestamp of last daily baseline payload save. Baselines saved once per day between 12:01 AM - 3:00 AM.';

backend/migrations/121_sales_analytics_views.sql

@@ -0,0 +1,383 @@
+-- Migration 121: Sales Analytics Materialized Views
+-- Pre-computed views for sales velocity, brand market share, and store performance
+
+-- ============================================================
+-- VIEW 1: Daily Sales Estimates (per product/store)
+-- Calculates delta between consecutive snapshots
+-- ============================================================
+CREATE MATERIALIZED VIEW IF NOT EXISTS mv_daily_sales_estimates AS
+WITH qty_deltas AS (
+  SELECT
+    dispensary_id,
+    product_id,
+    brand_name,
+    category,
+    DATE(captured_at) AS sale_date,
+    price_rec,
+    quantity_available,
+    LAG(quantity_available) OVER (
+      PARTITION BY dispensary_id, product_id
+      ORDER BY captured_at
+    ) AS prev_quantity
+  FROM inventory_snapshots
+  WHERE quantity_available IS NOT NULL
+    AND captured_at >= NOW() - INTERVAL '30 days'
+)
+SELECT
+  dispensary_id,
+  product_id,
+  brand_name,
+  category,
+  sale_date,
+  AVG(price_rec) AS avg_price,
+  SUM(GREATEST(0, COALESCE(prev_quantity, 0) - quantity_available)) AS units_sold,
+  SUM(GREATEST(0, quantity_available - COALESCE(prev_quantity, 0))) AS units_restocked,
+  SUM(GREATEST(0, COALESCE(prev_quantity, 0) - quantity_available) * COALESCE(price_rec, 0)) AS revenue_estimate,
+  COUNT(*) AS snapshot_count
+FROM qty_deltas
+WHERE prev_quantity IS NOT NULL
+GROUP BY dispensary_id, product_id, brand_name, category, sale_date;
+
+CREATE UNIQUE INDEX IF NOT EXISTS idx_mv_daily_sales_pk
+  ON mv_daily_sales_estimates(dispensary_id, product_id, sale_date);
+CREATE INDEX IF NOT EXISTS idx_mv_daily_sales_brand
+  ON mv_daily_sales_estimates(brand_name, sale_date);
+CREATE INDEX IF NOT EXISTS idx_mv_daily_sales_category
+  ON mv_daily_sales_estimates(category, sale_date);
+CREATE INDEX IF NOT EXISTS idx_mv_daily_sales_date
+  ON mv_daily_sales_estimates(sale_date DESC);
+
+
+-- ============================================================
+-- VIEW 2: Brand Market Share by State
+-- Weighted distribution across stores
+-- ============================================================
+CREATE MATERIALIZED VIEW IF NOT EXISTS mv_brand_market_share AS
+WITH brand_presence AS (
+  SELECT
+    sp.brand AS brand_name,
+    d.state AS state_code,
+    COUNT(DISTINCT sp.dispensary_id) AS stores_carrying,
+    COUNT(*) AS sku_count,
+    SUM(CASE WHEN sp.is_in_stock THEN 1 ELSE 0 END) AS in_stock_skus,
+    AVG(sp.price_rec) AS avg_price
+  FROM store_products sp
+  JOIN dispensaries d ON d.id = sp.dispensary_id
+  WHERE sp.brand IS NOT NULL
+    AND d.state IS NOT NULL
+  GROUP BY sp.brand, d.state
+),
+state_totals AS (
+  SELECT
+    d.state AS state_code,
+    COUNT(DISTINCT d.id) FILTER (WHERE d.crawl_enabled) AS total_stores
+  FROM dispensaries d
+  WHERE d.state IS NOT NULL
+  GROUP BY d.state
+)
+SELECT
+  bp.brand_name,
+  bp.state_code,
+  bp.stores_carrying,
+  st.total_stores,
+  ROUND(bp.stores_carrying::NUMERIC * 100 / NULLIF(st.total_stores, 0), 2) AS penetration_pct,
+  bp.sku_count,
+  bp.in_stock_skus,
+  bp.avg_price,
+  NOW() AS calculated_at
+FROM brand_presence bp
+JOIN state_totals st ON st.state_code = bp.state_code;
+
+CREATE UNIQUE INDEX IF NOT EXISTS idx_mv_brand_market_pk
+  ON mv_brand_market_share(brand_name, state_code);
+CREATE INDEX IF NOT EXISTS idx_mv_brand_market_state
+  ON mv_brand_market_share(state_code);
+CREATE INDEX IF NOT EXISTS idx_mv_brand_market_penetration
+  ON mv_brand_market_share(penetration_pct DESC);
+
+
+-- ============================================================
+-- VIEW 3: SKU Velocity (30-day rolling)
+-- Average daily units sold per SKU
+-- ============================================================
+CREATE MATERIALIZED VIEW IF NOT EXISTS mv_sku_velocity AS
+SELECT
+  dse.product_id,
+  dse.brand_name,
+  dse.category,
+  dse.dispensary_id,
+  d.name AS dispensary_name,
+  d.state AS state_code,
+  SUM(dse.units_sold) AS total_units_30d,
+  SUM(dse.revenue_estimate) AS total_revenue_30d,
+  COUNT(DISTINCT dse.sale_date) AS days_with_sales,
+  ROUND(SUM(dse.units_sold)::NUMERIC / NULLIF(COUNT(DISTINCT dse.sale_date), 0), 2) AS avg_daily_units,
+  AVG(dse.avg_price) AS avg_price,
+  CASE
+    WHEN SUM(dse.units_sold)::NUMERIC / NULLIF(COUNT(DISTINCT dse.sale_date), 0) >= 5 THEN 'hot'
+    WHEN SUM(dse.units_sold)::NUMERIC / NULLIF(COUNT(DISTINCT dse.sale_date), 0) >= 1 THEN 'steady'
+    WHEN SUM(dse.units_sold)::NUMERIC / NULLIF(COUNT(DISTINCT dse.sale_date), 0) >= 0.1 THEN 'slow'
+    ELSE 'stale'
+  END AS velocity_tier,
+  NOW() AS calculated_at
+FROM mv_daily_sales_estimates dse
+JOIN dispensaries d ON d.id = dse.dispensary_id
+WHERE dse.sale_date >= CURRENT_DATE - INTERVAL '30 days'
+GROUP BY dse.product_id, dse.brand_name, dse.category, dse.dispensary_id, d.name, d.state;
+
+CREATE UNIQUE INDEX IF NOT EXISTS idx_mv_sku_velocity_pk
+  ON mv_sku_velocity(dispensary_id, product_id);
+CREATE INDEX IF NOT EXISTS idx_mv_sku_velocity_brand
+  ON mv_sku_velocity(brand_name);
+CREATE INDEX IF NOT EXISTS idx_mv_sku_velocity_tier
+  ON mv_sku_velocity(velocity_tier);
+CREATE INDEX IF NOT EXISTS idx_mv_sku_velocity_state
+  ON mv_sku_velocity(state_code);
+CREATE INDEX IF NOT EXISTS idx_mv_sku_velocity_units
+  ON mv_sku_velocity(total_units_30d DESC);
+
+
+-- ============================================================
+-- VIEW 4: Store Performance Rankings
+-- Revenue estimates and brand diversity per store
+-- ============================================================
+CREATE MATERIALIZED VIEW IF NOT EXISTS mv_store_performance AS
+SELECT
+  d.id AS dispensary_id,
+  d.name AS dispensary_name,
+  d.city,
+  d.state AS state_code,
+  -- Revenue metrics from sales estimates
+  COALESCE(sales.total_revenue_30d, 0) AS total_revenue_30d,
+  COALESCE(sales.total_units_30d, 0) AS total_units_30d,
+  -- Inventory metrics
+  COUNT(DISTINCT sp.id) AS total_skus,
+  COUNT(DISTINCT sp.id) FILTER (WHERE sp.is_in_stock) AS in_stock_skus,
+  -- Brand diversity
+  COUNT(DISTINCT sp.brand) AS unique_brands,
+  COUNT(DISTINCT sp.category) AS unique_categories,
+  -- Pricing
+  AVG(sp.price_rec) AS avg_price,
+  -- Activity
+  MAX(sp.updated_at) AS last_updated,
+  NOW() AS calculated_at
+FROM dispensaries d
+LEFT JOIN store_products sp ON sp.dispensary_id = d.id
+LEFT JOIN (
+  SELECT
+    dispensary_id,
+    SUM(revenue_estimate) AS total_revenue_30d,
+    SUM(units_sold) AS total_units_30d
+  FROM mv_daily_sales_estimates
+  WHERE sale_date >= CURRENT_DATE - INTERVAL '30 days'
+  GROUP BY dispensary_id
+) sales ON sales.dispensary_id = d.id
+WHERE d.crawl_enabled = TRUE
+GROUP BY d.id, d.name, d.city, d.state, sales.total_revenue_30d, sales.total_units_30d;
+
+CREATE UNIQUE INDEX IF NOT EXISTS idx_mv_store_perf_pk
+  ON mv_store_performance(dispensary_id);
+CREATE INDEX IF NOT EXISTS idx_mv_store_perf_state
+  ON mv_store_performance(state_code);
+CREATE INDEX IF NOT EXISTS idx_mv_store_perf_revenue
+  ON mv_store_performance(total_revenue_30d DESC);
+
+
+-- ============================================================
+-- VIEW 5: Weekly Category Trends
+-- Category performance over time
+-- ============================================================
+CREATE MATERIALIZED VIEW IF NOT EXISTS mv_category_weekly_trends AS
+SELECT
+  dse.category,
+  d.state AS state_code,
+  DATE_TRUNC('week', dse.sale_date)::DATE AS week_start,
+  COUNT(DISTINCT dse.product_id) AS sku_count,
+  COUNT(DISTINCT dse.dispensary_id) AS store_count,
+  SUM(dse.units_sold) AS total_units,
+  SUM(dse.revenue_estimate) AS total_revenue,
+  AVG(dse.avg_price) AS avg_price,
+  NOW() AS calculated_at
+FROM mv_daily_sales_estimates dse
+JOIN dispensaries d ON d.id = dse.dispensary_id
+WHERE dse.category IS NOT NULL
+  AND dse.sale_date >= CURRENT_DATE - INTERVAL '90 days'
+GROUP BY dse.category, d.state, DATE_TRUNC('week', dse.sale_date);
+
+CREATE UNIQUE INDEX IF NOT EXISTS idx_mv_cat_weekly_pk
+  ON mv_category_weekly_trends(category, state_code, week_start);
+CREATE INDEX IF NOT EXISTS idx_mv_cat_weekly_state
+  ON mv_category_weekly_trends(state_code, week_start);
+CREATE INDEX IF NOT EXISTS idx_mv_cat_weekly_date
+  ON mv_category_weekly_trends(week_start DESC);
+
+
+-- ============================================================
+-- VIEW 6: Product Intelligence (Hoodie-style per-product metrics)
+-- Includes stock diff, days since OOS, days until stockout
+-- ============================================================
+CREATE MATERIALIZED VIEW IF NOT EXISTS mv_product_intelligence AS
+WITH
+-- Calculate stock diff over 120 days
+stock_diff AS (
+  SELECT
+    dispensary_id,
+    product_id,
+    -- Get oldest and newest quantity in last 120 days
+    FIRST_VALUE(quantity_available) OVER (
+      PARTITION BY dispensary_id, product_id
+      ORDER BY captured_at ASC
+      ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING
+    ) AS qty_120d_ago,
+    LAST_VALUE(quantity_available) OVER (
+      PARTITION BY dispensary_id, product_id
+      ORDER BY captured_at ASC
+      ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING
+    ) AS qty_current
+  FROM inventory_snapshots
+  WHERE captured_at >= NOW() - INTERVAL '120 days'
+),
+stock_diff_calc AS (
+  SELECT DISTINCT
+    dispensary_id,
+    product_id,
+    qty_current - COALESCE(qty_120d_ago, qty_current) AS stock_diff_120
+  FROM stock_diff
+),
+-- Get days since last OOS event
+last_oos AS (
+  SELECT
+    dispensary_id,
+    product_id,
+    MAX(detected_at) AS last_oos_date
+  FROM product_visibility_events
+  WHERE event_type = 'oos'
+  GROUP BY dispensary_id, product_id
+),
+-- Calculate avg daily units sold (from velocity view)
+velocity AS (
+  SELECT
+    dispensary_id,
+    product_id,
+    avg_daily_units
+  FROM mv_sku_velocity
+)
+SELECT
+  sp.dispensary_id,
+  d.name AS dispensary_name,
+  d.state AS state_code,
+  d.city,
+  sp.provider_product_id AS sku,
+  sp.name_raw AS product_name,
+  sp.brand_name_raw AS brand,
+  sp.category_raw AS category,
+  sp.is_in_stock,
+  sp.stock_status,
+  sp.stock_quantity,
+  sp.price_rec AS price,
+  sp.first_seen_at AS first_seen,
+  sp.last_seen_at AS last_seen,
+  -- Calculated fields
+  COALESCE(sd.stock_diff_120, 0) AS stock_diff_120,
+  CASE
+    WHEN lo.last_oos_date IS NOT NULL
+    THEN EXTRACT(DAY FROM NOW() - lo.last_oos_date)::INT
+    ELSE NULL
+  END AS days_since_oos,
+  -- Days until stockout = current stock / daily burn rate
+  CASE
+    WHEN v.avg_daily_units > 0 AND sp.stock_quantity > 0
+    THEN ROUND(sp.stock_quantity::NUMERIC / v.avg_daily_units)::INT
+    ELSE NULL
+  END AS days_until_stock_out,
+  v.avg_daily_units,
+  NOW() AS calculated_at
+FROM store_products sp
+JOIN dispensaries d ON d.id = sp.dispensary_id
+LEFT JOIN stock_diff_calc sd ON sd.dispensary_id = sp.dispensary_id
+  AND sd.product_id = sp.provider_product_id
+LEFT JOIN last_oos lo ON lo.dispensary_id = sp.dispensary_id
+  AND lo.product_id = sp.provider_product_id
+LEFT JOIN velocity v ON v.dispensary_id = sp.dispensary_id
+  AND v.product_id = sp.provider_product_id
+WHERE d.crawl_enabled = TRUE;
+
+CREATE UNIQUE INDEX IF NOT EXISTS idx_mv_prod_intel_pk
+  ON mv_product_intelligence(dispensary_id, sku);
+CREATE INDEX IF NOT EXISTS idx_mv_prod_intel_brand
+  ON mv_product_intelligence(brand);
+CREATE INDEX IF NOT EXISTS idx_mv_prod_intel_state
+  ON mv_product_intelligence(state_code);
+CREATE INDEX IF NOT EXISTS idx_mv_prod_intel_stock_out
+  ON mv_product_intelligence(days_until_stock_out ASC NULLS LAST);
+CREATE INDEX IF NOT EXISTS idx_mv_prod_intel_oos
+  ON mv_product_intelligence(days_since_oos DESC NULLS LAST);
+
+
+-- ============================================================
+-- REFRESH FUNCTION
+-- ============================================================
+CREATE OR REPLACE FUNCTION refresh_sales_analytics_views()
+RETURNS TABLE(view_name TEXT, rows_affected BIGINT) AS $$
+DECLARE
+  row_count BIGINT;
+BEGIN
+  -- Must refresh in dependency order:
+  -- 1. daily_sales (base view)
+  -- 2. sku_velocity (depends on daily_sales)
+  -- 3. product_intelligence (depends on sku_velocity)
+  -- 4. others (independent)
+
+  REFRESH MATERIALIZED VIEW CONCURRENTLY mv_daily_sales_estimates;
+  SELECT COUNT(*) INTO row_count FROM mv_daily_sales_estimates;
+  view_name := 'mv_daily_sales_estimates';
+  rows_affected := row_count;
+  RETURN NEXT;
+
+  REFRESH MATERIALIZED VIEW CONCURRENTLY mv_brand_market_share;
+  SELECT COUNT(*) INTO row_count FROM mv_brand_market_share;
+  view_name := 'mv_brand_market_share';
+  rows_affected := row_count;
+  RETURN NEXT;
+
+  REFRESH MATERIALIZED VIEW CONCURRENTLY mv_sku_velocity;
+  SELECT COUNT(*) INTO row_count FROM mv_sku_velocity;
+  view_name := 'mv_sku_velocity';
+  rows_affected := row_count;
+  RETURN NEXT;
+
+  REFRESH MATERIALIZED VIEW CONCURRENTLY mv_store_performance;
+  SELECT COUNT(*) INTO row_count FROM mv_store_performance;
+  view_name := 'mv_store_performance';
+  rows_affected := row_count;
+  RETURN NEXT;
+
+  REFRESH MATERIALIZED VIEW CONCURRENTLY mv_category_weekly_trends;
+  SELECT COUNT(*) INTO row_count FROM mv_category_weekly_trends;
+  view_name := 'mv_category_weekly_trends';
+  rows_affected := row_count;
+  RETURN NEXT;
+
+  -- Product intelligence depends on sku_velocity, so refresh last
+  REFRESH MATERIALIZED VIEW CONCURRENTLY mv_product_intelligence;
+  SELECT COUNT(*) INTO row_count FROM mv_product_intelligence;
+  view_name := 'mv_product_intelligence';
+  rows_affected := row_count;
+  RETURN NEXT;
+END;
+$$ LANGUAGE plpgsql;
+
+COMMENT ON FUNCTION refresh_sales_analytics_views IS
+  'Refresh all sales analytics materialized views. Call hourly via scheduler.';
+
+
+-- ============================================================
+-- INITIAL REFRESH (populate views)
+-- ============================================================
+-- Note: Initial refresh must be non-concurrent (no unique index yet populated)
+-- Run these manually after migration:
+-- REFRESH MATERIALIZED VIEW mv_daily_sales_estimates;
+-- REFRESH MATERIALIZED VIEW mv_brand_market_share;
+-- REFRESH MATERIALIZED VIEW mv_sku_velocity;
+-- REFRESH MATERIALIZED VIEW mv_store_performance;
+-- REFRESH MATERIALIZED VIEW mv_category_weekly_trends;

backend/migrations/122_market_intelligence_schema.sql

@@ -0,0 +1,359 @@
+-- Migration 122: Market Intelligence Schema
+-- Separate schema for external market data ingestion
+-- Supports product, brand, and dispensary data from third-party sources
+
+-- Create dedicated schema
+CREATE SCHEMA IF NOT EXISTS market_intel;
+
+-- ============================================================
+-- BRANDS: Brand/Company Intelligence
+-- ============================================================
+CREATE TABLE IF NOT EXISTS market_intel.brands (
+  id SERIAL PRIMARY KEY,
+
+  -- Identity
+  brand_name VARCHAR(255) NOT NULL,
+  parent_brand VARCHAR(255),
+  parent_company VARCHAR(255),
+  slug VARCHAR(255),
+  external_id VARCHAR(255) UNIQUE,  -- objectID from source
+
+  -- Details
+  brand_description TEXT,
+  brand_logo_url TEXT,
+  brand_url TEXT,
+  linkedin_url TEXT,
+
+  -- Presence
+  states JSONB DEFAULT '[]',  -- Array of state names
+  active_variants INTEGER DEFAULT 0,
+  all_variants INTEGER DEFAULT 0,
+
+  -- Metadata
+  source VARCHAR(50) DEFAULT 'external',
+  fetched_at TIMESTAMPTZ DEFAULT NOW(),
+  created_at TIMESTAMPTZ DEFAULT NOW(),
+  updated_at TIMESTAMPTZ DEFAULT NOW()
+);
+
+CREATE INDEX IF NOT EXISTS idx_brands_name ON market_intel.brands(brand_name);
+CREATE INDEX IF NOT EXISTS idx_brands_parent ON market_intel.brands(parent_brand);
+CREATE INDEX IF NOT EXISTS idx_brands_external ON market_intel.brands(external_id);
+CREATE INDEX IF NOT EXISTS idx_brands_states ON market_intel.brands USING GIN(states);
+
+
+-- ============================================================
+-- DISPENSARIES: Dispensary/Store Intelligence
+-- ============================================================
+CREATE TABLE IF NOT EXISTS market_intel.dispensaries (
+  id SERIAL PRIMARY KEY,
+
+  -- Identity
+  dispensary_name VARCHAR(255) NOT NULL,
+  dispensary_company_name VARCHAR(255),
+  dispensary_company_id VARCHAR(255),
+  slug VARCHAR(255),
+  external_id VARCHAR(255) UNIQUE,  -- objectID from source
+
+  -- Location
+  street_address VARCHAR(255),
+  city VARCHAR(100),
+  state VARCHAR(100),
+  postal_code VARCHAR(20),
+  county_name VARCHAR(100),
+  country_code VARCHAR(10) DEFAULT 'USA',
+  full_address TEXT,
+  latitude DECIMAL(10, 7),
+  longitude DECIMAL(10, 7),
+  timezone VARCHAR(50),
+  urbanicity VARCHAR(50),  -- Urban, Suburban, Rural
+
+  -- Contact
+  phone VARCHAR(50),
+  email VARCHAR(255),
+  website TEXT,
+  linkedin_url TEXT,
+
+  -- License
+  license_number VARCHAR(100),
+  license_type VARCHAR(100),
+
+  -- Store Type
+  is_medical BOOLEAN DEFAULT FALSE,
+  is_recreational BOOLEAN DEFAULT FALSE,
+  delivery_enabled BOOLEAN DEFAULT FALSE,
+  curbside_pickup BOOLEAN DEFAULT FALSE,
+  instore_pickup BOOLEAN DEFAULT FALSE,
+  location_type VARCHAR(50),  -- RETAIL, DELIVERY, etc.
+
+  -- Sales Estimates
+  estimated_daily_sales DECIMAL(12, 2),
+  estimated_sales DECIMAL(12, 2),
+  avg_daily_sales DECIMAL(12, 2),
+  state_sales_bucket INTEGER,
+
+  -- Customer Demographics
+  affluency JSONB DEFAULT '[]',  -- Array of affluency segments
+  age_skew JSONB DEFAULT '[]',   -- Array of age brackets
+  customer_segments JSONB DEFAULT '[]',  -- Array of segment names
+
+  -- Inventory Stats
+  menus_count INTEGER DEFAULT 0,
+  menus_count_med INTEGER DEFAULT 0,
+  menus_count_rec INTEGER DEFAULT 0,
+  parent_brands JSONB DEFAULT '[]',
+  brand_company_names JSONB DEFAULT '[]',
+
+  -- Business Info
+  banner VARCHAR(255),  -- Chain/banner name
+  business_type VARCHAR(50),  -- MSO, Independent, etc.
+  pos_system VARCHAR(100),
+  atm_presence BOOLEAN DEFAULT FALSE,
+  tax_included BOOLEAN DEFAULT FALSE,
+
+  -- Ratings
+  rating DECIMAL(3, 2),
+  reviews_count INTEGER DEFAULT 0,
+
+  -- Status
+  is_closed BOOLEAN DEFAULT FALSE,
+  open_date TIMESTAMPTZ,
+  last_updated_at TIMESTAMPTZ,
+
+  -- Media
+  logo_url TEXT,
+  cover_url TEXT,
+
+  -- Metadata
+  source VARCHAR(50) DEFAULT 'external',
+  fetched_at TIMESTAMPTZ DEFAULT NOW(),
+  created_at TIMESTAMPTZ DEFAULT NOW(),
+  updated_at TIMESTAMPTZ DEFAULT NOW()
+);
+
+CREATE INDEX IF NOT EXISTS idx_dispensaries_name ON market_intel.dispensaries(dispensary_name);
+CREATE INDEX IF NOT EXISTS idx_dispensaries_state ON market_intel.dispensaries(state);
+CREATE INDEX IF NOT EXISTS idx_dispensaries_city ON market_intel.dispensaries(city);
+CREATE INDEX IF NOT EXISTS idx_dispensaries_external ON market_intel.dispensaries(external_id);
+CREATE INDEX IF NOT EXISTS idx_dispensaries_banner ON market_intel.dispensaries(banner);
+CREATE INDEX IF NOT EXISTS idx_dispensaries_business_type ON market_intel.dispensaries(business_type);
+CREATE INDEX IF NOT EXISTS idx_dispensaries_geo ON market_intel.dispensaries(latitude, longitude);
+CREATE INDEX IF NOT EXISTS idx_dispensaries_segments ON market_intel.dispensaries USING GIN(customer_segments);
+
+
+-- ============================================================
+-- PRODUCTS: Product/SKU Intelligence
+-- ============================================================
+CREATE TABLE IF NOT EXISTS market_intel.products (
+  id SERIAL PRIMARY KEY,
+
+  -- Identity
+  name VARCHAR(500) NOT NULL,
+  brand VARCHAR(255),
+  brand_id VARCHAR(255),
+  brand_company_name VARCHAR(255),
+  parent_brand VARCHAR(255),
+  external_id VARCHAR(255) UNIQUE,  -- objectID from source
+  cm_id VARCHAR(100),  -- Canonical menu ID
+
+  -- Category Hierarchy
+  category_0 VARCHAR(100),  -- Top level: Flower, Edibles, Vapes
+  category_1 VARCHAR(255),  -- Mid level: Flower > Pre-Rolls
+  category_2 VARCHAR(500),  -- Detailed: Flower > Pre-Rolls > Singles
+
+  -- Cannabis Classification
+  cannabis_type VARCHAR(50),  -- SATIVA, INDICA, HYBRID
+  strain VARCHAR(255),
+  flavor VARCHAR(255),
+  pack_size VARCHAR(100),
+  description TEXT,
+
+  -- Cannabinoids
+  thc_mg DECIMAL(10, 2),
+  cbd_mg DECIMAL(10, 2),
+  percent_thc DECIMAL(5, 2),
+  percent_cbd DECIMAL(5, 2),
+
+  -- Dispensary Context (denormalized for query performance)
+  master_dispensary_name VARCHAR(255),
+  master_dispensary_id VARCHAR(255),
+  dispensary_count INTEGER DEFAULT 0,  -- How many stores carry this
+  d_state VARCHAR(100),
+  d_city VARCHAR(100),
+  d_banner VARCHAR(255),
+  d_business_type VARCHAR(50),
+  d_medical BOOLEAN,
+  d_recreational BOOLEAN,
+
+  -- Customer Demographics (from dispensary)
+  d_customer_segments JSONB DEFAULT '[]',
+  d_age_skew JSONB DEFAULT '[]',
+  d_affluency JSONB DEFAULT '[]',
+  d_urbanicity VARCHAR(50),
+
+  -- Stock Status
+  in_stock BOOLEAN DEFAULT TRUE,
+  last_seen_at DATE,
+  last_seen_at_ts BIGINT,
+
+  -- Media
+  img_url TEXT,
+  product_url TEXT,
+  menu_slug VARCHAR(500),
+
+  -- Geo
+  latitude DECIMAL(10, 7),
+  longitude DECIMAL(10, 7),
+
+  -- Metadata
+  source VARCHAR(50) DEFAULT 'external',
+  fetched_at TIMESTAMPTZ DEFAULT NOW(),
+  created_at TIMESTAMPTZ DEFAULT NOW(),
+  updated_at TIMESTAMPTZ DEFAULT NOW()
+);
+
+CREATE INDEX IF NOT EXISTS idx_products_name ON market_intel.products(name);
+CREATE INDEX IF NOT EXISTS idx_products_brand ON market_intel.products(brand);
+CREATE INDEX IF NOT EXISTS idx_products_external ON market_intel.products(external_id);
+CREATE INDEX IF NOT EXISTS idx_products_category ON market_intel.products(category_0, category_1);
+CREATE INDEX IF NOT EXISTS idx_products_cannabis_type ON market_intel.products(cannabis_type);
+CREATE INDEX IF NOT EXISTS idx_products_strain ON market_intel.products(strain);
+CREATE INDEX IF NOT EXISTS idx_products_state ON market_intel.products(d_state);
+CREATE INDEX IF NOT EXISTS idx_products_in_stock ON market_intel.products(in_stock);
+CREATE INDEX IF NOT EXISTS idx_products_dispensary_count ON market_intel.products(dispensary_count DESC);
+CREATE INDEX IF NOT EXISTS idx_products_segments ON market_intel.products USING GIN(d_customer_segments);
+
+
+-- ============================================================
+-- PRODUCT_VARIANTS: Variant-Level Data (Pricing, Stock)
+-- ============================================================
+CREATE TABLE IF NOT EXISTS market_intel.product_variants (
+  id SERIAL PRIMARY KEY,
+  product_id INTEGER REFERENCES market_intel.products(id) ON DELETE CASCADE,
+
+  -- Identity
+  variant_id VARCHAR(255) NOT NULL,
+  pos_sku VARCHAR(255),
+  pos_product_id VARCHAR(255),
+  pos_system VARCHAR(100),
+
+  -- Pricing
+  actual_price DECIMAL(10, 2),
+  original_price DECIMAL(10, 2),
+  discounted_price DECIMAL(10, 2),
+
+  -- Presentation
+  product_presentation VARCHAR(255),  -- "100.00 mg", "3.5g", etc.
+  quantity DECIMAL(10, 2),
+  unit VARCHAR(50),  -- mg, g, oz, each
+
+  -- Availability
+  is_medical BOOLEAN DEFAULT FALSE,
+  is_recreational BOOLEAN DEFAULT FALSE,
+  is_active BOOLEAN DEFAULT TRUE,
+
+  -- Stock Intelligence
+  stock_status VARCHAR(50),  -- In Stock, Low Stock, Out of Stock
+  stock_diff_120 DECIMAL(10, 2),  -- 120-day stock change
+  days_since_oos INTEGER,
+  days_until_stock_out INTEGER,
+
+  -- Timestamps
+  first_seen_at_ts BIGINT,
+  first_seen_at TIMESTAMPTZ,
+  last_seen_at DATE,
+
+  -- Metadata
+  fetched_at TIMESTAMPTZ DEFAULT NOW(),
+  created_at TIMESTAMPTZ DEFAULT NOW(),
+  updated_at TIMESTAMPTZ DEFAULT NOW(),
+
+  UNIQUE(product_id, variant_id)
+);
+
+CREATE INDEX IF NOT EXISTS idx_variants_product ON market_intel.product_variants(product_id);
+CREATE INDEX IF NOT EXISTS idx_variants_sku ON market_intel.product_variants(pos_sku);
+CREATE INDEX IF NOT EXISTS idx_variants_stock_status ON market_intel.product_variants(stock_status);
+CREATE INDEX IF NOT EXISTS idx_variants_price ON market_intel.product_variants(actual_price);
+CREATE INDEX IF NOT EXISTS idx_variants_days_out ON market_intel.product_variants(days_until_stock_out);
+
+
+-- ============================================================
+-- FETCH_LOG: Track data fetches
+-- ============================================================
+CREATE TABLE IF NOT EXISTS market_intel.fetch_log (
+  id SERIAL PRIMARY KEY,
+  fetch_type VARCHAR(50) NOT NULL,  -- brands, dispensaries, products
+  state_code VARCHAR(10),
+  query_params JSONB,
+  records_fetched INTEGER DEFAULT 0,
+  records_inserted INTEGER DEFAULT 0,
+  records_updated INTEGER DEFAULT 0,
+  duration_ms INTEGER,
+  error_message TEXT,
+  started_at TIMESTAMPTZ DEFAULT NOW(),
+  completed_at TIMESTAMPTZ
+);
+
+CREATE INDEX IF NOT EXISTS idx_fetch_log_type ON market_intel.fetch_log(fetch_type);
+CREATE INDEX IF NOT EXISTS idx_fetch_log_state ON market_intel.fetch_log(state_code);
+CREATE INDEX IF NOT EXISTS idx_fetch_log_started ON market_intel.fetch_log(started_at DESC);
+
+
+-- ============================================================
+-- HELPER VIEWS
+-- ============================================================
+
+-- Brand market presence summary
+CREATE OR REPLACE VIEW market_intel.v_brand_presence AS
+SELECT
+  b.brand_name,
+  b.parent_company,
+  b.active_variants,
+  b.all_variants,
+  jsonb_array_length(b.states) as state_count,
+  b.states,
+  b.fetched_at
+FROM market_intel.brands b
+ORDER BY b.active_variants DESC;
+
+-- Dispensary sales rankings by state
+CREATE OR REPLACE VIEW market_intel.v_dispensary_rankings AS
+SELECT
+  d.dispensary_name,
+  d.city,
+  d.state,
+  d.banner,
+  d.business_type,
+  d.estimated_daily_sales,
+  d.menus_count,
+  d.is_medical,
+  d.is_recreational,
+  d.customer_segments,
+  RANK() OVER (PARTITION BY d.state ORDER BY d.estimated_daily_sales DESC NULLS LAST) as state_rank
+FROM market_intel.dispensaries d
+WHERE d.is_closed = FALSE;
+
+-- Product distribution by brand and state
+CREATE OR REPLACE VIEW market_intel.v_product_distribution AS
+SELECT
+  p.brand,
+  p.d_state as state,
+  p.category_0 as category,
+  COUNT(*) as product_count,
+  COUNT(*) FILTER (WHERE p.in_stock) as in_stock_count,
+  AVG(p.dispensary_count) as avg_store_count,
+  COUNT(DISTINCT p.master_dispensary_id) as unique_stores
+FROM market_intel.products p
+GROUP BY p.brand, p.d_state, p.category_0;
+
+
+-- ============================================================
+-- COMMENTS
+-- ============================================================
+COMMENT ON SCHEMA market_intel IS 'Market intelligence data from external sources';
+COMMENT ON TABLE market_intel.brands IS 'Brand/company data with multi-state presence';
+COMMENT ON TABLE market_intel.dispensaries IS 'Dispensary data with sales estimates and demographics';
+COMMENT ON TABLE market_intel.products IS 'Product/SKU data with cannabinoid and category info';
+COMMENT ON TABLE market_intel.product_variants IS 'Variant-level pricing and stock data';
+COMMENT ON TABLE market_intel.fetch_log IS 'Log of data fetches for monitoring';

backend/migrations/123_extract_provider_fields.sql

@@ -0,0 +1,159 @@
+-- Migration 123: Extract unmapped fields from provider_data
+-- These fields exist in our crawl payloads but weren't being stored in columns
+
+-- ============================================================
+-- ADD NEW COLUMNS TO store_products
+-- ============================================================
+
+-- Cannabis classification (SATIVA, INDICA, HYBRID, CBD)
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS cannabis_type VARCHAR(50);
+
+-- Canonical IDs from POS systems
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS canonical_strain_id VARCHAR(100);
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS canonical_vendor_id VARCHAR(100);
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS canonical_brand_id VARCHAR(100);
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS canonical_category_id VARCHAR(100);
+
+-- Lab results
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS lab_result_url TEXT;
+
+-- Flavors (extracted from JSONB to text array for easier querying)
+ALTER TABLE store_products ADD COLUMN IF NOT EXISTS flavors_list TEXT[];
+
+-- ============================================================
+-- BACKFILL FROM provider_data
+-- ============================================================
+
+-- Backfill cannabis_type from classification
+UPDATE store_products
+SET cannabis_type = CASE
+  WHEN provider_data->>'classification' IN ('HYBRID', 'H') THEN 'HYBRID'
+  WHEN provider_data->>'classification' IN ('INDICA', 'I') THEN 'INDICA'
+  WHEN provider_data->>'classification' IN ('SATIVA', 'S') THEN 'SATIVA'
+  WHEN provider_data->>'classification' = 'I/S' THEN 'INDICA_DOMINANT'
+  WHEN provider_data->>'classification' = 'S/I' THEN 'SATIVA_DOMINANT'
+  WHEN provider_data->>'classification' = 'CBD' THEN 'CBD'
+  ELSE provider_data->>'classification'
+END
+WHERE provider_data->>'classification' IS NOT NULL
+  AND cannabis_type IS NULL;
+
+-- Also backfill from strain_type if cannabis_type still null
+UPDATE store_products
+SET cannabis_type = CASE
+  WHEN strain_type ILIKE '%indica%hybrid%' OR strain_type ILIKE '%hybrid%indica%' THEN 'INDICA_DOMINANT'
+  WHEN strain_type ILIKE '%sativa%hybrid%' OR strain_type ILIKE '%hybrid%sativa%' THEN 'SATIVA_DOMINANT'
+  WHEN strain_type ILIKE '%indica%' THEN 'INDICA'
+  WHEN strain_type ILIKE '%sativa%' THEN 'SATIVA'
+  WHEN strain_type ILIKE '%hybrid%' THEN 'HYBRID'
+  WHEN strain_type ILIKE '%cbd%' THEN 'CBD'
+  ELSE NULL
+END
+WHERE strain_type IS NOT NULL
+  AND cannabis_type IS NULL;
+
+-- Backfill canonical IDs from POSMetaData
+UPDATE store_products
+SET
+  canonical_strain_id = provider_data->'POSMetaData'->>'canonicalStrainId',
+  canonical_vendor_id = provider_data->'POSMetaData'->>'canonicalVendorId',
+  canonical_brand_id = provider_data->'POSMetaData'->>'canonicalBrandId',
+  canonical_category_id = provider_data->'POSMetaData'->>'canonicalCategoryId'
+WHERE provider_data->'POSMetaData' IS NOT NULL
+  AND canonical_strain_id IS NULL;
+
+-- Backfill lab result URLs
+UPDATE store_products
+SET lab_result_url = provider_data->'POSMetaData'->>'canonicalLabResultUrl'
+WHERE provider_data->'POSMetaData'->>'canonicalLabResultUrl' IS NOT NULL
+  AND lab_result_url IS NULL;
+
+-- ============================================================
+-- INDEXES
+-- ============================================================
+
+CREATE INDEX IF NOT EXISTS idx_store_products_cannabis_type ON store_products(cannabis_type);
+CREATE INDEX IF NOT EXISTS idx_store_products_vendor_id ON store_products(canonical_vendor_id);
+CREATE INDEX IF NOT EXISTS idx_store_products_strain_id ON store_products(canonical_strain_id);
+
+-- ============================================================
+-- ADD MSO FLAG TO DISPENSARIES
+-- ============================================================
+
+-- Multi-State Operator flag (calculated from chain presence in multiple states)
+ALTER TABLE dispensaries ADD COLUMN IF NOT EXISTS is_mso BOOLEAN DEFAULT FALSE;
+
+-- Update MSO flag based on chain presence in multiple states
+WITH mso_chains AS (
+  SELECT chain_id
+  FROM dispensaries
+  WHERE chain_id IS NOT NULL
+  GROUP BY chain_id
+  HAVING COUNT(DISTINCT state) > 1
+)
+UPDATE dispensaries d
+SET is_mso = TRUE
+WHERE d.chain_id IN (SELECT chain_id FROM mso_chains);
+
+-- Index for MSO queries
+CREATE INDEX IF NOT EXISTS idx_dispensaries_is_mso ON dispensaries(is_mso) WHERE is_mso = TRUE;
+
+-- ============================================================
+-- PRODUCT DISTRIBUTION VIEW
+-- ============================================================
+
+-- View: How many stores carry each product (by brand + canonical name)
+CREATE OR REPLACE VIEW v_product_distribution AS
+SELECT
+  sp.brand_name_raw as brand,
+  sp.c_name as product_canonical_name,
+  COUNT(DISTINCT sp.dispensary_id) as store_count,
+  COUNT(DISTINCT d.state) as state_count,
+  ARRAY_AGG(DISTINCT d.state) as states,
+  AVG(sp.price_rec) as avg_price,
+  MIN(sp.price_rec) as min_price,
+  MAX(sp.price_rec) as max_price
+FROM store_products sp
+JOIN dispensaries d ON d.id = sp.dispensary_id
+WHERE sp.c_name IS NOT NULL
+  AND sp.brand_name_raw IS NOT NULL
+  AND sp.is_in_stock = TRUE
+GROUP BY sp.brand_name_raw, sp.c_name
+HAVING COUNT(DISTINCT sp.dispensary_id) > 1
+ORDER BY store_count DESC;
+
+-- ============================================================
+-- MSO SUMMARY VIEW
+-- ============================================================
+
+CREATE OR REPLACE VIEW v_mso_summary AS
+SELECT
+  c.name as chain_name,
+  COUNT(DISTINCT d.id) as store_count,
+  COUNT(DISTINCT d.state) as state_count,
+  ARRAY_AGG(DISTINCT d.state ORDER BY d.state) as states,
+  SUM(d.product_count) as total_products,
+  TRUE as is_mso
+FROM dispensaries d
+JOIN chains c ON c.id = d.chain_id
+WHERE d.chain_id IN (
+  SELECT chain_id
+  FROM dispensaries
+  WHERE chain_id IS NOT NULL
+  GROUP BY chain_id
+  HAVING COUNT(DISTINCT state) > 1
+)
+GROUP BY c.id, c.name
+ORDER BY state_count DESC, store_count DESC;
+
+-- ============================================================
+-- COMMENTS
+-- ============================================================
+
+COMMENT ON COLUMN store_products.cannabis_type IS 'Normalized cannabis classification: SATIVA, INDICA, HYBRID, INDICA_DOMINANT, SATIVA_DOMINANT, CBD';
+COMMENT ON COLUMN store_products.canonical_strain_id IS 'POS system strain identifier for cross-store matching';
+COMMENT ON COLUMN store_products.canonical_vendor_id IS 'POS system vendor/supplier identifier';
+COMMENT ON COLUMN store_products.lab_result_url IS 'Link to Certificate of Analysis / lab test results';
+COMMENT ON COLUMN dispensaries.is_mso IS 'Multi-State Operator: chain operates in 2+ states';
+COMMENT ON VIEW v_product_distribution IS 'Shows how many stores carry each product for distribution analysis';
+COMMENT ON VIEW v_mso_summary IS 'Summary of multi-state operator chains';

backend/migrations/124_timescaledb_snapshots.sql

@@ -0,0 +1,73 @@
+-- Migration 124: Convert inventory_snapshots to TimescaleDB hypertable
+-- Requires: CREATE EXTENSION timescaledb; (run after installing TimescaleDB)
+
+-- ============================================================
+-- STEP 1: Enable TimescaleDB extension
+-- ============================================================
+CREATE EXTENSION IF NOT EXISTS timescaledb;
+
+-- ============================================================
+-- STEP 2: Convert to hypertable
+-- ============================================================
+-- Note: Table must have a time column and no foreign key constraints
+
+-- First, drop any foreign keys if they exist
+ALTER TABLE inventory_snapshots DROP CONSTRAINT IF EXISTS inventory_snapshots_dispensary_id_fkey;
+
+-- Convert to hypertable, partitioned by captured_at (1 day chunks)
+SELECT create_hypertable(
+  'inventory_snapshots',
+  'captured_at',
+  chunk_time_interval => INTERVAL '1 day',
+  if_not_exists => TRUE,
+  migrate_data => TRUE
+);
+
+-- ============================================================
+-- STEP 3: Enable compression
+-- ============================================================
+-- Compress by dispensary_id and product_id (common query patterns)
+ALTER TABLE inventory_snapshots SET (
+  timescaledb.compress,
+  timescaledb.compress_segmentby = 'dispensary_id, product_id',
+  timescaledb.compress_orderby = 'captured_at DESC'
+);
+
+-- ============================================================
+-- STEP 4: Compression policy (compress chunks older than 1 day)
+-- ============================================================
+SELECT add_compression_policy('inventory_snapshots', INTERVAL '1 day');
+
+-- ============================================================
+-- STEP 5: Retention policy (optional - drop chunks older than 90 days)
+-- ============================================================
+-- Uncomment if you want automatic cleanup:
+-- SELECT add_retention_policy('inventory_snapshots', INTERVAL '90 days');
+
+-- ============================================================
+-- STEP 6: Optimize indexes for time-series queries
+-- ============================================================
+-- TimescaleDB automatically creates time-based indexes
+-- Add composite index for common queries
+CREATE INDEX IF NOT EXISTS idx_snapshots_disp_prod_time
+  ON inventory_snapshots (dispensary_id, product_id, captured_at DESC);
+
+-- ============================================================
+-- VERIFICATION QUERIES (run after migration)
+-- ============================================================
+-- Check hypertable status:
+-- SELECT * FROM timescaledb_information.hypertables WHERE hypertable_name = 'inventory_snapshots';
+
+-- Check compression status:
+-- SELECT * FROM timescaledb_information.compression_settings WHERE hypertable_name = 'inventory_snapshots';
+
+-- Check chunk sizes:
+-- SELECT chunk_name, pg_size_pretty(before_compression_total_bytes) as before,
+--        pg_size_pretty(after_compression_total_bytes) as after,
+--        round(100 - (after_compression_total_bytes::numeric / before_compression_total_bytes * 100), 1) as compression_pct
+-- FROM chunk_compression_stats('inventory_snapshots');
+
+-- ============================================================
+-- COMMENTS
+-- ============================================================
+COMMENT ON TABLE inventory_snapshots IS 'TimescaleDB hypertable for inventory time-series data. Compressed after 1 day.';

backend/migrations/125_delta_only_snapshots.sql

@@ -0,0 +1,402 @@
+-- Migration 125: Delta-only inventory snapshots
+-- Only store a row when something meaningful changes
+-- Revenue calculated as: effective_price × qty_sold
+
+-- ============================================================
+-- ADD DELTA TRACKING COLUMNS
+-- ============================================================
+
+-- Previous values (to show what changed)
+ALTER TABLE inventory_snapshots ADD COLUMN IF NOT EXISTS prev_quantity INTEGER;
+ALTER TABLE inventory_snapshots ADD COLUMN IF NOT EXISTS prev_price_rec DECIMAL(10,2);
+ALTER TABLE inventory_snapshots ADD COLUMN IF NOT EXISTS prev_price_med DECIMAL(10,2);
+ALTER TABLE inventory_snapshots ADD COLUMN IF NOT EXISTS prev_status VARCHAR(50);
+
+-- Calculated deltas
+ALTER TABLE inventory_snapshots ADD COLUMN IF NOT EXISTS qty_delta INTEGER; -- negative = sold, positive = restocked
+ALTER TABLE inventory_snapshots ADD COLUMN IF NOT EXISTS price_delta DECIMAL(10,2);
+
+-- Change type flags
+ALTER TABLE inventory_snapshots ADD COLUMN IF NOT EXISTS change_type VARCHAR(50); -- 'sale', 'restock', 'price_change', 'oos', 'back_in_stock'
+
+-- ============================================================
+-- INDEX FOR CHANGE TYPE QUERIES
+-- ============================================================
+CREATE INDEX IF NOT EXISTS idx_snapshots_change_type ON inventory_snapshots(change_type);
+CREATE INDEX IF NOT EXISTS idx_snapshots_qty_delta ON inventory_snapshots(qty_delta) WHERE qty_delta != 0;
+
+-- ============================================================
+-- VIEW: Latest product state (for delta comparison)
+-- ============================================================
+CREATE OR REPLACE VIEW v_product_latest_state AS
+SELECT DISTINCT ON (dispensary_id, product_id)
+  dispensary_id,
+  product_id,
+  quantity_available,
+  price_rec,
+  price_med,
+  status,
+  captured_at
+FROM inventory_snapshots
+ORDER BY dispensary_id, product_id, captured_at DESC;
+
+-- ============================================================
+-- FUNCTION: Check if product state changed
+-- ============================================================
+CREATE OR REPLACE FUNCTION should_capture_snapshot(
+  p_dispensary_id INTEGER,
+  p_product_id TEXT,
+  p_quantity INTEGER,
+  p_price_rec DECIMAL,
+  p_price_med DECIMAL,
+  p_status VARCHAR
+) RETURNS TABLE (
+  should_capture BOOLEAN,
+  prev_quantity INTEGER,
+  prev_price_rec DECIMAL,
+  prev_price_med DECIMAL,
+  prev_status VARCHAR,
+  qty_delta INTEGER,
+  price_delta DECIMAL,
+  change_type VARCHAR
+) AS $$
+DECLARE
+  v_prev RECORD;
+BEGIN
+  -- Get previous state
+  SELECT
+    ls.quantity_available,
+    ls.price_rec,
+    ls.price_med,
+    ls.status
+  INTO v_prev
+  FROM v_product_latest_state ls
+  WHERE ls.dispensary_id = p_dispensary_id
+    AND ls.product_id = p_product_id;
+
+  -- First time seeing this product
+  IF NOT FOUND THEN
+    RETURN QUERY SELECT
+      TRUE,
+      NULL::INTEGER,
+      NULL::DECIMAL,
+      NULL::DECIMAL,
+      NULL::VARCHAR,
+      NULL::INTEGER,
+      NULL::DECIMAL,
+      'new_product'::VARCHAR;
+    RETURN;
+  END IF;
+
+  -- Check for changes
+  IF v_prev.quantity_available IS DISTINCT FROM p_quantity
+     OR v_prev.price_rec IS DISTINCT FROM p_price_rec
+     OR v_prev.price_med IS DISTINCT FROM p_price_med
+     OR v_prev.status IS DISTINCT FROM p_status THEN
+
+    RETURN QUERY SELECT
+      TRUE,
+      v_prev.quantity_available,
+      v_prev.price_rec,
+      v_prev.price_med,
+      v_prev.status,
+      COALESCE(p_quantity, 0) - COALESCE(v_prev.quantity_available, 0),
+      COALESCE(p_price_rec, 0) - COALESCE(v_prev.price_rec, 0),
+      CASE
+        WHEN COALESCE(p_quantity, 0) < COALESCE(v_prev.quantity_available, 0) THEN 'sale'
+        WHEN COALESCE(p_quantity, 0) > COALESCE(v_prev.quantity_available, 0) THEN 'restock'
+        WHEN p_quantity = 0 AND v_prev.quantity_available > 0 THEN 'oos'
+        WHEN p_quantity > 0 AND v_prev.quantity_available = 0 THEN 'back_in_stock'
+        WHEN p_price_rec IS DISTINCT FROM v_prev.price_rec THEN 'price_change'
+        ELSE 'status_change'
+      END;
+    RETURN;
+  END IF;
+
+  -- No change
+  RETURN QUERY SELECT
+    FALSE,
+    NULL::INTEGER,
+    NULL::DECIMAL,
+    NULL::DECIMAL,
+    NULL::VARCHAR,
+    NULL::INTEGER,
+    NULL::DECIMAL,
+    NULL::VARCHAR;
+END;
+$$ LANGUAGE plpgsql;
+
+-- ============================================================
+-- REVENUE CALCULATION COLUMNS
+-- ============================================================
+-- Effective prices (sale price if on special, otherwise regular)
+ALTER TABLE inventory_snapshots ADD COLUMN IF NOT EXISTS effective_price_rec DECIMAL(10,2);
+ALTER TABLE inventory_snapshots ADD COLUMN IF NOT EXISTS effective_price_med DECIMAL(10,2);
+ALTER TABLE inventory_snapshots ADD COLUMN IF NOT EXISTS is_on_special BOOLEAN DEFAULT FALSE;
+
+-- Revenue by market type
+ALTER TABLE inventory_snapshots ADD COLUMN IF NOT EXISTS revenue_rec DECIMAL(10,2);
+ALTER TABLE inventory_snapshots ADD COLUMN IF NOT EXISTS revenue_med DECIMAL(10,2);
+
+-- Time between snapshots (for velocity calc)
+ALTER TABLE inventory_snapshots ADD COLUMN IF NOT EXISTS time_since_last_snapshot INTERVAL;
+ALTER TABLE inventory_snapshots ADD COLUMN IF NOT EXISTS hours_since_last DECIMAL(10,2);
+
+-- ============================================================
+-- VIEW: Hourly Sales Velocity
+-- ============================================================
+CREATE OR REPLACE VIEW v_hourly_sales AS
+SELECT
+  dispensary_id,
+  DATE(captured_at) as sale_date,
+  EXTRACT(HOUR FROM captured_at) as sale_hour,
+  COUNT(*) FILTER (WHERE qty_delta < 0) as transactions,
+  SUM(ABS(qty_delta)) FILTER (WHERE qty_delta < 0) as units_sold,
+  SUM(revenue_estimate) FILTER (WHERE qty_delta < 0) as revenue,
+  COUNT(DISTINCT product_id) FILTER (WHERE qty_delta < 0) as unique_products_sold
+FROM inventory_snapshots
+WHERE change_type = 'sale'
+GROUP BY dispensary_id, DATE(captured_at), EXTRACT(HOUR FROM captured_at);
+
+-- ============================================================
+-- VIEW: Daily Sales by Store
+-- ============================================================
+CREATE OR REPLACE VIEW v_daily_store_sales AS
+SELECT
+  s.dispensary_id,
+  d.name as store_name,
+  d.state,
+  DATE(s.captured_at) as sale_date,
+  SUM(ABS(s.qty_delta)) as units_sold,
+  SUM(s.revenue_estimate) as revenue,
+  COUNT(*) as sale_events,
+  COUNT(DISTINCT s.product_id) as unique_products
+FROM inventory_snapshots s
+JOIN dispensaries d ON d.id = s.dispensary_id
+WHERE s.change_type = 'sale'
+GROUP BY s.dispensary_id, d.name, d.state, DATE(s.captured_at);
+
+-- ============================================================
+-- VIEW: Daily Sales by Brand
+-- ============================================================
+CREATE OR REPLACE VIEW v_daily_brand_sales AS
+SELECT
+  s.brand_name,
+  d.state,
+  DATE(s.captured_at) as sale_date,
+  SUM(ABS(s.qty_delta)) as units_sold,
+  SUM(s.revenue_estimate) as revenue,
+  COUNT(DISTINCT s.dispensary_id) as stores_with_sales,
+  COUNT(DISTINCT s.product_id) as unique_skus_sold
+FROM inventory_snapshots s
+JOIN dispensaries d ON d.id = s.dispensary_id
+WHERE s.change_type = 'sale'
+  AND s.brand_name IS NOT NULL
+GROUP BY s.brand_name, d.state, DATE(s.captured_at);
+
+-- ============================================================
+-- VIEW: Product Velocity Rankings
+-- ============================================================
+CREATE OR REPLACE VIEW v_product_velocity AS
+SELECT
+  s.product_id,
+  s.brand_name,
+  s.category,
+  s.dispensary_id,
+  d.name as store_name,
+  d.state,
+  SUM(ABS(s.qty_delta)) as units_sold_30d,
+  SUM(s.revenue_estimate) as revenue_30d,
+  COUNT(*) as sale_events,
+  ROUND(SUM(ABS(s.qty_delta))::NUMERIC / NULLIF(COUNT(DISTINCT DATE(s.captured_at)), 0), 2) as avg_daily_units,
+  ROUND(SUM(s.revenue_estimate) / NULLIF(COUNT(DISTINCT DATE(s.captured_at)), 0), 2) as avg_daily_revenue,
+  CASE
+    WHEN SUM(ABS(s.qty_delta)) / NULLIF(COUNT(DISTINCT DATE(s.captured_at)), 0) >= 10 THEN 'hot'
+    WHEN SUM(ABS(s.qty_delta)) / NULLIF(COUNT(DISTINCT DATE(s.captured_at)), 0) >= 3 THEN 'steady'
+    WHEN SUM(ABS(s.qty_delta)) / NULLIF(COUNT(DISTINCT DATE(s.captured_at)), 0) >= 1 THEN 'slow'
+    ELSE 'stale'
+  END as velocity_tier
+FROM inventory_snapshots s
+JOIN dispensaries d ON d.id = s.dispensary_id
+WHERE s.change_type = 'sale'
+  AND s.captured_at >= NOW() - INTERVAL '30 days'
+GROUP BY s.product_id, s.brand_name, s.category, s.dispensary_id, d.name, d.state;
+
+-- ============================================================
+-- VIEW: Busiest Hours by Store
+-- ============================================================
+CREATE OR REPLACE VIEW v_busiest_hours AS
+SELECT
+  dispensary_id,
+  sale_hour,
+  AVG(units_sold) as avg_units_per_hour,
+  AVG(revenue) as avg_revenue_per_hour,
+  SUM(units_sold) as total_units,
+  SUM(revenue) as total_revenue,
+  COUNT(*) as days_with_data,
+  RANK() OVER (PARTITION BY dispensary_id ORDER BY AVG(revenue) DESC) as hour_rank
+FROM v_hourly_sales
+GROUP BY dispensary_id, sale_hour;
+
+-- ============================================================
+-- VIEW: Promotion Effectiveness (compare sale vs non-sale prices)
+-- ============================================================
+CREATE OR REPLACE VIEW v_promotion_effectiveness AS
+SELECT
+  s.dispensary_id,
+  d.name as store_name,
+  s.product_id,
+  s.brand_name,
+  DATE(s.captured_at) as sale_date,
+  SUM(ABS(s.qty_delta)) FILTER (WHERE s.price_rec < s.prev_price_rec) as units_on_discount,
+  SUM(ABS(s.qty_delta)) FILTER (WHERE s.price_rec >= COALESCE(s.prev_price_rec, s.price_rec)) as units_full_price,
+  SUM(s.revenue_estimate) FILTER (WHERE s.price_rec < s.prev_price_rec) as revenue_discounted,
+  SUM(s.revenue_estimate) FILTER (WHERE s.price_rec >= COALESCE(s.prev_price_rec, s.price_rec)) as revenue_full_price
+FROM inventory_snapshots s
+JOIN dispensaries d ON d.id = s.dispensary_id
+WHERE s.change_type = 'sale'
+GROUP BY s.dispensary_id, d.name, s.product_id, s.brand_name, DATE(s.captured_at);
+
+-- ============================================================
+-- COMMENTS
+-- ============================================================
+COMMENT ON COLUMN inventory_snapshots.qty_delta IS 'Quantity change: negative=sold, positive=restocked';
+COMMENT ON COLUMN inventory_snapshots.revenue_estimate IS 'Estimated revenue: ABS(qty_delta) * price_rec when qty_delta < 0';
+COMMENT ON COLUMN inventory_snapshots.change_type IS 'Type of change: sale, restock, price_change, oos, back_in_stock, new_product';
+COMMENT ON FUNCTION should_capture_snapshot IS 'Returns whether a snapshot should be captured and delta values';
+COMMENT ON VIEW v_hourly_sales IS 'Sales aggregated by hour - find busiest times';
+COMMENT ON VIEW v_daily_store_sales IS 'Daily revenue by store';
+COMMENT ON VIEW v_daily_brand_sales IS 'Daily brand performance by state';
+COMMENT ON VIEW v_product_velocity IS 'Product sales velocity rankings (hot/steady/slow/stale)';
+COMMENT ON VIEW v_busiest_hours IS 'Rank hours by sales volume per store';
+
+-- ============================================================
+-- VIEW: Days Until Stock Out (Predictive)
+-- ============================================================
+CREATE OR REPLACE VIEW v_stock_out_prediction AS
+WITH velocity AS (
+  SELECT
+    dispensary_id,
+    product_id,
+    brand_name,
+    -- Average units sold per day (last 7 days)
+    ROUND(SUM(ABS(qty_delta))::NUMERIC / NULLIF(COUNT(DISTINCT DATE(captured_at)), 0), 2) as daily_velocity,
+    -- Hours between sales
+    AVG(hours_since_last) FILTER (WHERE qty_delta < 0) as avg_hours_between_sales
+  FROM inventory_snapshots
+  WHERE change_type = 'sale'
+    AND captured_at >= NOW() - INTERVAL '7 days'
+  GROUP BY dispensary_id, product_id, brand_name
+),
+current_stock AS (
+  SELECT DISTINCT ON (dispensary_id, product_id)
+    dispensary_id,
+    product_id,
+    quantity_available as current_qty,
+    captured_at as last_seen
+  FROM inventory_snapshots
+  ORDER BY dispensary_id, product_id, captured_at DESC
+)
+SELECT
+  cs.dispensary_id,
+  d.name as store_name,
+  cs.product_id,
+  v.brand_name,
+  cs.current_qty,
+  v.daily_velocity,
+  CASE
+    WHEN v.daily_velocity > 0 THEN ROUND(cs.current_qty / v.daily_velocity, 1)
+    ELSE NULL
+  END as days_until_stock_out,
+  CASE
+    WHEN v.daily_velocity > 0 AND cs.current_qty / v.daily_velocity <= 3 THEN 'critical'
+    WHEN v.daily_velocity > 0 AND cs.current_qty / v.daily_velocity <= 7 THEN 'low'
+    WHEN v.daily_velocity > 0 AND cs.current_qty / v.daily_velocity <= 14 THEN 'moderate'
+    ELSE 'healthy'
+  END as stock_health,
+  cs.last_seen
+FROM current_stock cs
+JOIN velocity v ON v.dispensary_id = cs.dispensary_id AND v.product_id = cs.product_id
+JOIN dispensaries d ON d.id = cs.dispensary_id
+WHERE cs.current_qty > 0
+  AND v.daily_velocity > 0;
+
+-- ============================================================
+-- VIEW: Days Since OOS (for products currently out of stock)
+-- ============================================================
+CREATE OR REPLACE VIEW v_days_since_oos AS
+SELECT
+  s.dispensary_id,
+  d.name as store_name,
+  s.product_id,
+  s.brand_name,
+  s.captured_at as went_oos_at,
+  EXTRACT(EPOCH FROM (NOW() - s.captured_at)) / 86400 as days_since_oos,
+  s.prev_quantity as last_known_qty
+FROM inventory_snapshots s
+JOIN dispensaries d ON d.id = s.dispensary_id
+WHERE s.change_type = 'oos'
+  AND NOT EXISTS (
+    -- No back_in_stock event after this OOS
+    SELECT 1 FROM inventory_snapshots s2
+    WHERE s2.dispensary_id = s.dispensary_id
+      AND s2.product_id = s.product_id
+      AND s2.change_type = 'back_in_stock'
+      AND s2.captured_at > s.captured_at
+  );
+
+-- ============================================================
+-- VIEW: Brand Variant Counts (track brand growth)
+-- ============================================================
+CREATE OR REPLACE VIEW v_brand_variants AS
+SELECT
+  sp.brand_name_raw as brand_name,
+  d.state,
+  COUNT(DISTINCT sp.id) as total_variants,
+  COUNT(DISTINCT sp.id) FILTER (WHERE sp.is_in_stock = TRUE) as active_variants,
+  COUNT(DISTINCT sp.id) FILTER (WHERE sp.is_in_stock = FALSE) as inactive_variants,
+  COUNT(DISTINCT sp.dispensary_id) as stores_carrying,
+  COUNT(DISTINCT sp.category_raw) as categories,
+  MIN(sp.first_seen_at) as brand_first_seen,
+  MAX(sp.last_seen_at) as brand_last_seen
+FROM store_products sp
+JOIN dispensaries d ON d.id = sp.dispensary_id
+WHERE sp.brand_name_raw IS NOT NULL
+GROUP BY sp.brand_name_raw, d.state;
+
+-- ============================================================
+-- VIEW: Brand Growth (compare variant counts over time)
+-- ============================================================
+CREATE OR REPLACE VIEW v_brand_growth AS
+WITH weekly_counts AS (
+  SELECT
+    brand_name_raw as brand_name,
+    DATE_TRUNC('week', last_seen_at) as week,
+    COUNT(DISTINCT id) as variant_count
+  FROM store_products
+  WHERE brand_name_raw IS NOT NULL
+    AND last_seen_at >= NOW() - INTERVAL '90 days'
+  GROUP BY brand_name_raw, DATE_TRUNC('week', last_seen_at)
+)
+SELECT
+  w1.brand_name,
+  w1.week as current_week,
+  w1.variant_count as current_variants,
+  w2.variant_count as prev_week_variants,
+  w1.variant_count - COALESCE(w2.variant_count, 0) as variant_change,
+  CASE
+    WHEN w2.variant_count IS NULL THEN 'new'
+    WHEN w1.variant_count > w2.variant_count THEN 'growing'
+    WHEN w1.variant_count < w2.variant_count THEN 'declining'
+    ELSE 'stable'
+  END as growth_status
+FROM weekly_counts w1
+LEFT JOIN weekly_counts w2
+  ON w2.brand_name = w1.brand_name
+  AND w2.week = w1.week - INTERVAL '1 week'
+ORDER BY w1.brand_name, w1.week DESC;
+
+COMMENT ON VIEW v_stock_out_prediction IS 'Predict days until stock out based on velocity';
+COMMENT ON VIEW v_days_since_oos IS 'Products currently OOS and how long they have been out';
+COMMENT ON VIEW v_brand_variants IS 'Active vs inactive SKU counts per brand per state';
+COMMENT ON VIEW v_brand_growth IS 'Week-over-week brand variant growth tracking';

backend/migrations/126_az_high_frequency.sql

@@ -0,0 +1,53 @@
+-- Migration 126: Set AZ stores to 5-minute high-frequency crawls
+-- Other states default to 60-minute (1 hour) intervals
+
+-- ============================================================
+-- SET AZ STORES TO 5-MINUTE INTERVALS (with 3-min jitter)
+-- ============================================================
+-- Base interval: 5 minutes
+-- Jitter: +/- 3 minutes (so 2-8 minute effective range)
+UPDATE dispensaries
+SET
+  crawl_interval_minutes = 5,
+  next_crawl_at = NOW() + (RANDOM() * INTERVAL '5 minutes') -- Stagger initial crawls
+WHERE state = 'AZ'
+  AND crawl_enabled = TRUE;
+
+-- ============================================================
+-- SET OTHER STATES TO 60-MINUTE INTERVALS (with 3-min jitter)
+-- ============================================================
+UPDATE dispensaries
+SET
+  crawl_interval_minutes = 60,
+  next_crawl_at = NOW() + (RANDOM() * INTERVAL '60 minutes') -- Stagger initial crawls
+WHERE state != 'AZ'
+  AND crawl_enabled = TRUE
+  AND crawl_interval_minutes IS NULL;
+
+-- ============================================================
+-- VERIFY RESULTS
+-- ============================================================
+-- SELECT state, crawl_interval_minutes, COUNT(*)
+-- FROM dispensaries
+-- WHERE crawl_enabled = TRUE
+-- GROUP BY state, crawl_interval_minutes
+-- ORDER BY state;
+
+-- ============================================================
+-- CREATE VIEW FOR MONITORING CRAWL LOAD
+-- ============================================================
+CREATE OR REPLACE VIEW v_crawl_load AS
+SELECT
+  state,
+  crawl_interval_minutes,
+  COUNT(*) as store_count,
+  -- Crawls per hour = stores * (60 / interval)
+  ROUND(COUNT(*) * (60.0 / COALESCE(crawl_interval_minutes, 60))) as crawls_per_hour,
+  -- Assuming 30 sec per crawl, workers needed = crawls_per_hour / 120
+  ROUND(COUNT(*) * (60.0 / COALESCE(crawl_interval_minutes, 60)) / 120, 1) as workers_needed
+FROM dispensaries
+WHERE crawl_enabled = TRUE
+GROUP BY state, crawl_interval_minutes
+ORDER BY crawls_per_hour DESC;
+
+COMMENT ON VIEW v_crawl_load IS 'Monitor crawl load by state and interval';

backend/migrations/127_fix_worker_task_limit.sql

@@ -0,0 +1,164 @@
+-- Migration 127: Fix worker task concurrency limit
+-- Problem: claim_task function checks session_task_count but never increments it
+-- Solution: Increment on claim, decrement on complete/fail/release
+
+-- =============================================================================
+-- STEP 1: Set max tasks to 5 for all workers
+-- =============================================================================
+UPDATE worker_registry SET session_max_tasks = 5;
+
+-- Set default to 5 for new workers
+ALTER TABLE worker_registry ALTER COLUMN session_max_tasks SET DEFAULT 5;
+
+-- =============================================================================
+-- STEP 2: Reset all session_task_count to match actual active tasks
+-- =============================================================================
+UPDATE worker_registry wr SET session_task_count = (
+  SELECT COUNT(*) FROM worker_tasks wt
+  WHERE wt.worker_id = wr.worker_id
+    AND wt.status IN ('claimed', 'running')
+);
+
+-- =============================================================================
+-- STEP 3: Update claim_task function to increment session_task_count
+-- =============================================================================
+CREATE OR REPLACE FUNCTION claim_task(
+  p_role VARCHAR(50),
+  p_worker_id VARCHAR(100),
+  p_curl_passed BOOLEAN DEFAULT TRUE,
+  p_http_passed BOOLEAN DEFAULT FALSE
+) RETURNS worker_tasks AS $$
+DECLARE
+  claimed_task worker_tasks;
+  worker_state VARCHAR(2);
+  session_valid BOOLEAN;
+  session_tasks INT;
+  max_tasks INT;
+BEGIN
+  -- Get worker's current geo session info
+  SELECT
+    current_state,
+    session_task_count,
+    session_max_tasks,
+    (geo_session_started_at IS NOT NULL AND geo_session_started_at > NOW() - INTERVAL '60 minutes')
+  INTO worker_state, session_tasks, max_tasks, session_valid
+  FROM worker_registry
+  WHERE worker_id = p_worker_id;
+
+  -- Check if worker has reached max concurrent tasks (default 5)
+  IF session_tasks >= COALESCE(max_tasks, 5) THEN
+    RETURN NULL;
+  END IF;
+
+  -- If no valid geo session, or session expired, worker can't claim tasks
+  -- Worker must re-qualify first
+  IF worker_state IS NULL OR NOT session_valid THEN
+    RETURN NULL;
+  END IF;
+
+  -- Claim task matching worker's state
+  UPDATE worker_tasks
+  SET
+    status = 'claimed',
+    worker_id = p_worker_id,
+    claimed_at = NOW(),
+    updated_at = NOW()
+  WHERE id = (
+    SELECT wt.id FROM worker_tasks wt
+    JOIN dispensaries d ON wt.dispensary_id = d.id
+    WHERE wt.role = p_role
+      AND wt.status = 'pending'
+      AND (wt.scheduled_for IS NULL OR wt.scheduled_for <= NOW())
+      -- GEO FILTER: Task's dispensary must match worker's state
+      AND d.state = worker_state
+      -- Method compatibility: worker must have passed the required preflight
+      AND (
+        wt.method IS NULL  -- No preference, any worker can claim
+        OR (wt.method = 'curl' AND p_curl_passed = TRUE)
+        OR (wt.method = 'http' AND p_http_passed = TRUE)
+      )
+      -- Exclude stores that already have an active task
+      AND (wt.dispensary_id IS NULL OR wt.dispensary_id NOT IN (
+        SELECT dispensary_id FROM worker_tasks
+        WHERE status IN ('claimed', 'running')
+        AND dispensary_id IS NOT NULL
+        AND dispensary_id != wt.dispensary_id
+      ))
+    ORDER BY wt.priority DESC, wt.created_at ASC
+    LIMIT 1
+    FOR UPDATE SKIP LOCKED
+  )
+  RETURNING * INTO claimed_task;
+
+  -- INCREMENT session_task_count if we claimed a task
+  IF claimed_task.id IS NOT NULL THEN
+    UPDATE worker_registry
+    SET session_task_count = session_task_count + 1
+    WHERE worker_id = p_worker_id;
+  END IF;
+
+  RETURN claimed_task;
+END;
+$$ LANGUAGE plpgsql;
+
+-- =============================================================================
+-- STEP 4: Create trigger to decrement on task completion/failure/release
+-- =============================================================================
+CREATE OR REPLACE FUNCTION decrement_worker_task_count()
+RETURNS TRIGGER AS $$
+BEGIN
+  -- Only decrement when task was assigned to a worker and is now complete/released
+  IF OLD.worker_id IS NOT NULL AND OLD.status IN ('claimed', 'running') THEN
+    -- Task completed/failed/released - decrement count
+    IF NEW.status IN ('pending', 'completed', 'failed') OR NEW.worker_id IS NULL THEN
+      UPDATE worker_registry
+      SET session_task_count = GREATEST(0, session_task_count - 1)
+      WHERE worker_id = OLD.worker_id;
+    END IF;
+  END IF;
+
+  RETURN NEW;
+END;
+$$ LANGUAGE plpgsql;
+
+-- Drop existing trigger if any
+DROP TRIGGER IF EXISTS trg_decrement_worker_task_count ON worker_tasks;
+
+-- Create trigger on UPDATE (status change or worker_id cleared)
+CREATE TRIGGER trg_decrement_worker_task_count
+AFTER UPDATE ON worker_tasks
+FOR EACH ROW
+EXECUTE FUNCTION decrement_worker_task_count();
+
+-- Also handle DELETE (completed tasks are deleted from pool)
+CREATE OR REPLACE FUNCTION decrement_worker_task_count_delete()
+RETURNS TRIGGER AS $$
+BEGIN
+  IF OLD.worker_id IS NOT NULL AND OLD.status IN ('claimed', 'running') THEN
+    UPDATE worker_registry
+    SET session_task_count = GREATEST(0, session_task_count - 1)
+    WHERE worker_id = OLD.worker_id;
+  END IF;
+  RETURN OLD;
+END;
+$$ LANGUAGE plpgsql;
+
+DROP TRIGGER IF EXISTS trg_decrement_worker_task_count_delete ON worker_tasks;
+
+CREATE TRIGGER trg_decrement_worker_task_count_delete
+AFTER DELETE ON worker_tasks
+FOR EACH ROW
+EXECUTE FUNCTION decrement_worker_task_count_delete();
+
+-- =============================================================================
+-- STEP 5: Verify current state
+-- =============================================================================
+SELECT
+  wr.worker_id,
+  wr.friendly_name,
+  wr.session_task_count,
+  wr.session_max_tasks,
+  (SELECT COUNT(*) FROM worker_tasks wt WHERE wt.worker_id = wr.worker_id AND wt.status IN ('claimed', 'running')) as actual_count
+FROM worker_registry wr
+WHERE wr.status = 'active'
+ORDER BY wr.friendly_name;

backend/migrations/128_pool_config.sql

@@ -0,0 +1,109 @@
+-- Migration 128: Pool configuration table
+-- Controls whether workers can claim tasks from the pool
+
+CREATE TABLE IF NOT EXISTS pool_config (
+  id SERIAL PRIMARY KEY,
+  pool_open BOOLEAN NOT NULL DEFAULT true,
+  closed_reason TEXT,
+  closed_at TIMESTAMPTZ,
+  closed_by VARCHAR(100),
+  opened_at TIMESTAMPTZ DEFAULT NOW(),
+  created_at TIMESTAMPTZ DEFAULT NOW(),
+  updated_at TIMESTAMPTZ DEFAULT NOW()
+);
+
+-- Insert default config (pool open)
+INSERT INTO pool_config (pool_open, opened_at)
+VALUES (true, NOW())
+ON CONFLICT DO NOTHING;
+
+-- Update claim_task function to check pool status
+CREATE OR REPLACE FUNCTION claim_task(
+  p_role VARCHAR(50),
+  p_worker_id VARCHAR(100),
+  p_curl_passed BOOLEAN DEFAULT TRUE,
+  p_http_passed BOOLEAN DEFAULT FALSE
+) RETURNS worker_tasks AS $$
+DECLARE
+  claimed_task worker_tasks;
+  worker_state VARCHAR(2);
+  session_valid BOOLEAN;
+  session_tasks INT;
+  max_tasks INT;
+  is_pool_open BOOLEAN;
+BEGIN
+  -- Check if pool is open
+  SELECT pool_open INTO is_pool_open FROM pool_config LIMIT 1;
+  IF NOT COALESCE(is_pool_open, true) THEN
+    RETURN NULL;  -- Pool is closed, no claiming allowed
+  END IF;
+
+  -- Get worker's current geo session info
+  SELECT
+    current_state,
+    session_task_count,
+    session_max_tasks,
+    (geo_session_started_at IS NOT NULL AND geo_session_started_at > NOW() - INTERVAL '60 minutes')
+  INTO worker_state, session_tasks, max_tasks, session_valid
+  FROM worker_registry
+  WHERE worker_id = p_worker_id;
+
+  -- Check if worker has reached max concurrent tasks (default 5)
+  IF session_tasks >= COALESCE(max_tasks, 5) THEN
+    RETURN NULL;
+  END IF;
+
+  -- If no valid geo session, or session expired, worker can't claim tasks
+  -- Worker must re-qualify first
+  IF worker_state IS NULL OR NOT session_valid THEN
+    RETURN NULL;
+  END IF;
+
+  -- Claim task matching worker's state
+  UPDATE worker_tasks
+  SET
+    status = 'claimed',
+    worker_id = p_worker_id,
+    claimed_at = NOW(),
+    updated_at = NOW()
+  WHERE id = (
+    SELECT wt.id FROM worker_tasks wt
+    JOIN dispensaries d ON wt.dispensary_id = d.id
+    WHERE wt.role = p_role
+      AND wt.status = 'pending'
+      AND (wt.scheduled_for IS NULL OR wt.scheduled_for <= NOW())
+      -- GEO FILTER: Task's dispensary must match worker's state
+      AND d.state = worker_state
+      -- Method compatibility: worker must have passed the required preflight
+      AND (
+        wt.method IS NULL  -- No preference, any worker can claim
+        OR (wt.method = 'curl' AND p_curl_passed = TRUE)
+        OR (wt.method = 'http' AND p_http_passed = TRUE)
+      )
+      -- Exclude stores that already have an active task
+      AND (wt.dispensary_id IS NULL OR wt.dispensary_id NOT IN (
+        SELECT dispensary_id FROM worker_tasks
+        WHERE status IN ('claimed', 'running')
+        AND dispensary_id IS NOT NULL
+        AND dispensary_id != wt.dispensary_id
+      ))
+    ORDER BY wt.priority DESC, wt.created_at ASC
+    LIMIT 1
+    FOR UPDATE SKIP LOCKED
+  )
+  RETURNING * INTO claimed_task;
+
+  -- INCREMENT session_task_count if we claimed a task
+  IF claimed_task.id IS NOT NULL THEN
+    UPDATE worker_registry
+    SET session_task_count = session_task_count + 1
+    WHERE worker_id = p_worker_id;
+  END IF;
+
+  RETURN claimed_task;
+END;
+$$ LANGUAGE plpgsql;
+
+-- Verify
+SELECT 'pool_config table created' as status;
+SELECT * FROM pool_config;

backend/migrations/129_claim_tasks_for_geo.sql

@@ -0,0 +1,60 @@
+-- Migration 129: Claim tasks for specific geo
+-- Used after worker gets IP to claim more tasks for same geo
+
+-- Function: Claim up to N tasks for a SPECIFIC geo (state/city)
+-- Different from claim_tasks_batch which picks the geo with most tasks
+CREATE OR REPLACE FUNCTION claim_tasks_batch_for_geo(
+  p_worker_id VARCHAR(255),
+  p_max_tasks INTEGER DEFAULT 4,
+  p_state_code VARCHAR(2),
+  p_city VARCHAR(100) DEFAULT NULL,
+  p_role VARCHAR(50) DEFAULT NULL
+) RETURNS TABLE (
+  task_id INTEGER,
+  role VARCHAR(50),
+  dispensary_id INTEGER,
+  dispensary_name VARCHAR(255),
+  city VARCHAR(100),
+  state_code VARCHAR(2),
+  platform VARCHAR(50),
+  method VARCHAR(20)
+) AS $$
+BEGIN
+  -- Claim up to p_max_tasks for the specified geo
+  RETURN QUERY
+  WITH claimed AS (
+    UPDATE worker_tasks t SET
+      status = 'claimed',
+      worker_id = p_worker_id,
+      claimed_at = NOW()
+    FROM (
+      SELECT t2.id
+      FROM worker_tasks t2
+      JOIN dispensaries d ON t2.dispensary_id = d.id
+      WHERE t2.status = 'pending'
+        AND d.state = p_state_code
+        AND (p_city IS NULL OR d.city = p_city)
+        AND (p_role IS NULL OR t2.role = p_role)
+      ORDER BY t2.priority DESC, t2.created_at ASC
+      FOR UPDATE SKIP LOCKED
+      LIMIT p_max_tasks
+    ) sub
+    WHERE t.id = sub.id
+    RETURNING t.id, t.role, t.dispensary_id, t.method
+  )
+  SELECT
+    c.id as task_id,
+    c.role,
+    c.dispensary_id,
+    d.name as dispensary_name,
+    d.city,
+    d.state as state_code,
+    d.platform,
+    c.method
+  FROM claimed c
+  JOIN dispensaries d ON c.dispensary_id = d.id;
+END;
+$$ LANGUAGE plpgsql;
+
+-- Verify
+SELECT 'claim_tasks_batch_for_geo function created' as status;

backend/migrations/130_hoodie_comparison_reports.sql

@@ -0,0 +1,49 @@
+-- Hoodie Comparison Reports
+-- Stores delta results from comparing Hoodie data against CannaIQ
+-- Raw Hoodie data stays remote (proxy only) - we only store comparison results
+
+CREATE TABLE IF NOT EXISTS hoodie_comparison_reports (
+  id SERIAL PRIMARY KEY,
+  report_type VARCHAR(50) NOT NULL,  -- 'dispensaries', 'brands', 'products'
+  state VARCHAR(50) NOT NULL,
+
+  -- Counts
+  hoodie_total INT NOT NULL DEFAULT 0,
+  cannaiq_total INT NOT NULL DEFAULT 0,
+  in_both INT NOT NULL DEFAULT 0,
+  hoodie_only INT NOT NULL DEFAULT 0,
+  cannaiq_only INT NOT NULL DEFAULT 0,
+
+  -- Delta details (JSONB for flexibility)
+  hoodie_only_items JSONB DEFAULT '[]',   -- Items in Hoodie but not CannaIQ
+  cannaiq_only_items JSONB DEFAULT '[]',  -- Items in CannaIQ but not Hoodie
+  matched_items JSONB DEFAULT '[]',       -- Items in both (with any differences)
+
+  -- Metadata
+  created_at TIMESTAMPTZ DEFAULT NOW(),
+  duration_ms INT,  -- How long the comparison took
+  error TEXT        -- Any errors during comparison
+);
+
+-- Index for querying latest reports
+CREATE INDEX idx_hoodie_reports_type_state ON hoodie_comparison_reports(report_type, state, created_at DESC);
+CREATE INDEX idx_hoodie_reports_created ON hoodie_comparison_reports(created_at DESC);
+
+-- View for latest report per type/state
+CREATE OR REPLACE VIEW v_hoodie_latest_reports AS
+SELECT DISTINCT ON (report_type, state)
+  id,
+  report_type,
+  state,
+  hoodie_total,
+  cannaiq_total,
+  in_both,
+  hoodie_only,
+  cannaiq_only,
+  created_at,
+  duration_ms
+FROM hoodie_comparison_reports
+WHERE error IS NULL
+ORDER BY report_type, state, created_at DESC;
+
+COMMENT ON TABLE hoodie_comparison_reports IS 'Stores comparison results between Hoodie and CannaIQ data. Raw Hoodie data stays remote.';

backend/migrations/130_worker_badge.sql

@@ -0,0 +1,53 @@
+-- Migration 130: Worker qualification badge
+-- Session-scoped badge showing worker qualification status
+
+-- Add badge column to worker_registry
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS badge VARCHAR(20) DEFAULT NULL;
+
+-- Add qualified_at timestamp
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS qualified_at TIMESTAMPTZ DEFAULT NULL;
+
+-- Add current_session_id to link worker to their active session
+ALTER TABLE worker_registry
+ADD COLUMN IF NOT EXISTS current_session_id INTEGER DEFAULT NULL;
+
+-- Badge values:
+-- 'gold' = preflight passed, actively qualified with valid session
+-- NULL = not qualified (no active session or session expired)
+
+-- Function: Set worker badge to gold when qualified
+CREATE OR REPLACE FUNCTION set_worker_qualified(
+  p_worker_id VARCHAR(255),
+  p_session_id INTEGER
+) RETURNS BOOLEAN AS $$
+BEGIN
+  UPDATE worker_registry
+  SET badge = 'gold',
+      qualified_at = NOW(),
+      current_session_id = p_session_id
+  WHERE worker_id = p_worker_id;
+  RETURN FOUND;
+END;
+$$ LANGUAGE plpgsql;
+
+-- Function: Clear worker badge when session ends
+CREATE OR REPLACE FUNCTION clear_worker_badge(p_worker_id VARCHAR(255))
+RETURNS BOOLEAN AS $$
+BEGIN
+  UPDATE worker_registry
+  SET badge = NULL,
+      qualified_at = NULL,
+      current_session_id = NULL
+  WHERE worker_id = p_worker_id;
+  RETURN FOUND;
+END;
+$$ LANGUAGE plpgsql;
+
+-- Index for finding qualified workers
+CREATE INDEX IF NOT EXISTS idx_worker_registry_badge
+  ON worker_registry(badge) WHERE badge IS NOT NULL;
+
+-- Verify
+SELECT 'worker_registry badge column added' as status;

backend/migrations/131_normalize_brand.sql

@@ -0,0 +1,21 @@
+-- Migration: 131_normalize_brand
+-- Purpose: Add normalize_brand() function for fuzzy brand matching across dispensaries
+-- Used by Cannabrands integration to match brand names regardless of spelling variations
+
+-- Function to normalize brand names for matching
+-- "Aloha TymeMachine" → "alohatymemachine"
+-- "ALOHA TYME MACHINE" → "alohatymemachine"
+-- "Aloha Tyme Machine" → "alohatymemachine"
+CREATE OR REPLACE FUNCTION normalize_brand(name TEXT)
+RETURNS TEXT AS $$
+  SELECT LOWER(REGEXP_REPLACE(COALESCE(name, ''), '[^a-zA-Z0-9]', '', 'g'))
+$$ LANGUAGE SQL IMMUTABLE PARALLEL SAFE;
+
+-- Create functional index for efficient lookups
+-- This allows queries like: WHERE normalize_brand(brand_name_raw) = 'alohatymemachine'
+CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_store_products_brand_normalized
+ON store_products (normalize_brand(brand_name_raw));
+
+-- Also index on snapshots table for historical queries
+CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_store_product_snapshots_brand_normalized
+ON store_product_snapshots (normalize_brand(brand_name_raw));