chore: remove cloudflare + docker + legacy migration scripts

Phase 7 cleanup of the Vercel + Upstash consolidation plan:

- delete wrangler.toml, Dockerfile, docker-compose{,.dev}.yml,
  scripts/migrate-atlas-to-upstash.js (one-shot migration done)
- drop wrangler + mongodb devDeps and migrate* npm scripts;
  regenerate package-lock.json (-70 packages)
- prune CF/Wrangler/Atlas-export entries from .gitignore + .vercelignore
- drop MONGODB_URI from .env.deploy.example
- rewrite README for Vercel + Upstash architecture
- refresh stale Cloudflare comments in src/{logger,models,repository}

.env.deploy.example

@@ -16,7 +16,3 @@ WORKER_URL=https://store-scraper-bot.vercel.app/api/webhook
 UPSTASH_REDIS_REST_URL=
 UPSTASH_REDIS_REST_TOKEN=
 KEY_PREFIX=store-scraper-bot:
-
-# Legacy Java-bot MongoDB Atlas — required for one-shot Phase 5 migration only.
-# After successful cutover, remove this line.
-MONGODB_URI=

.gitignore

@@ -10,13 +10,6 @@ pnpm-debug.log*
 .env.deploy
 .dev.vars
 
-# Cloudflare Workers
-.tmp-deploy/
-.wrangler/
-
-# Atlas → KV migration intermediate (contains exported bot state)
-scripts/.atlas-export.json
-
 # Editor / IDE
 # .idea/
 # .vscode/

.vercelignore

@@ -2,14 +2,8 @@ plans/
 docs/
 *.md
 LICENSE
-Dockerfile
-docker-compose.yml
-docker-compose.dev.yml
-wrangler.toml
 .env*
 scripts/check-secret-leaks.js
-scripts/migrate-atlas-to-upstash.js
-scripts/migrate-atlas-to-kv.js
 scripts/register-webhook.js
 .git
 .github

Dockerfile

@@ -1,13 +0,0 @@
-FROM node:20-alpine
-
-RUN apk --no-cache add tzdata ca-certificates
-
-WORKDIR /app
-
-COPY package.json ./
-RUN npm install --omit=dev
-
-COPY src ./src
-
-# Bot uses long polling — no ports exposed.
-CMD ["node", "src/index.js"]

README.md

@@ -1,6 +1,7 @@
 # store-scraper-bot
 
 JavaScript (Node.js) implementation. Ports [java-store-scraper-bot](https://github.com/tiennm99/java-store-scraper-bot).
+Runs on Vercel serverless functions with Upstash Redis as the data store.
 
 > ⚠️ **Preview / unstable — use at your own risk.**
 > This port was produced largely with AI assistance and has **not** been tested
@@ -11,88 +12,95 @@ The Java version remains the reference implementation.
 
 ## Status
 
-- Mongo schema matches Java/Go (collections: `common`, `group`, `apple_app`,
+- Upstash Redis schema mirrors the Java/Go Mongo layout: keys `admin`,
-  `google_app`; string `_id`; `class` discriminator).
+  `group:{chatId}`, `apple:{appId}`, `google:{appId}` (last two TTL'd via Redis
+  `EX`). Multi-tenant isolation via `KEY_PREFIX` (default `store-scraper-bot:`).
 - Telegram command identifiers match Java exactly: `/info`, `/addgroup`,
   `/delgroup`, `/listgroup`, `/addapple`, `/delapple`, `/addgoogle`,
   `/delgoogle`, `/listapp`, `/checkapp`, `/checkappscore`, `/rawappleapp`,
   `/rawgoogleapp`.
-- HTML parse mode; weekend-silent daily report; configurable API cache (default 10 min).
+- HTML parse mode; weekend-silent daily report; configurable upstream cache (default 10 min).
+- Inlined `app-store-scraper` + `google-play-scraper` (no external scraper service).
 
 ## Requirements
 
 - Node.js 20+ (uses built-in `fetch`)
-- MongoDB 4.4+
+- Vercel account (Hobby plan / free tier is enough)
+- Upstash Redis database (free tier; sign up at upstash.com or via Vercel Marketplace)
 
 ## Configuration
 
-See `.env.example`:
+Vercel env vars:
 
 | Name | Notes |
 |---|---|
 | `TELEGRAM_BOT_TOKEN` | Telegram bot token (required) |
 | `TELEGRAM_BOT_USERNAME` | Bot username (required) |
-| `MONGODB_CONNECTION_STRING` | Preferred (Java parity); falls back to `MONGO_URI` |
+| `TELEGRAM_WEBHOOK_SECRET` | ≥32 chars random; verifies inbound webhook calls |
-| `MONGO_DATABASE` | Optional; inferred from URI path if omitted |
 | `ADMIN_IDS` | Comma-separated Telegram user IDs (required) |
+| `UPSTASH_REDIS_REST_URL` | Upstash REST endpoint (or `KV_REST_API_URL` from Vercel Marketplace integration) |
+| `UPSTASH_REDIS_REST_TOKEN` | Upstash REST token (or `KV_REST_API_TOKEN` fallback) |
+| `KEY_PREFIX` | Namespace for all Redis keys (default `store-scraper-bot:`) |
+| `CRON_SECRET` | ≥32 chars random; required by Vercel Cron handler |
 | `ENV` | `DEVELOPMENT` or `PRODUCTION` |
 | `SOURCE_COMMIT` | Optional; shown on startup |
 | `APP_CACHE_SECONDS` | Cache TTL for upstream API responses (default 600) |
 | `NUM_DAYS_WARNING_NOT_UPDATED` | Threshold for daily warning (default 30) |
 | `SCHEDULE_CHECK_APP_TIME` | Cron expression in Vietnam timezone (default `0 7 * * *`) |
 
+Operator-only `.env.deploy` (used by `npm run register`) — see `.env.deploy.example`.
+
 ## Run
 
+Local dev:
+
 ```sh
 npm install
-cp .env.example .env   # then edit credentials
+vercel link            # link to your Vercel project
-npm start
+vercel env pull .env.local
+npm run dev            # vercel dev
 ```
 
-Or via Docker Compose:
+Deploy:
 
 ```sh
-docker compose up --build
+npm run deploy         # vercel deploy --prod && register webhook
 ```
 
-## Migrating from MongoDB Atlas (one-time)
+`npm run register` re-points the Telegram webhook at the URL in `.env.deploy:WORKER_URL`.
-
-If you're moving an existing bot off Atlas to Cloudflare KV:
-
-```sh
-# 1. Make sure .env has MONGODB_URI pointing at the Atlas cluster (read-only is fine).
-npm install                    # pulls the mongodb devDep needed by the migration script
-npm run migrate                # writes scripts/.atlas-export.json (admin + groups)
-# Optionally also migrate cached app entries with their TTL preserved:
-#   npm run migrate -- --include-cache
-npm run migrate:bulk           # uploads the JSON to the production KV namespace
-rm scripts/.atlas-export.json  # contains your bot state — delete after success
-```
-
-Cache collections (`apple_app`, `google_app`) are skipped by default since they auto-rebuild from upstream APIs within `APP_CACHE_SECONDS`. Re-running the migration is idempotent.
 
 ## Project Layout
 
 ```
+api/
+├── webhook.js          # Telegram webhook entry (Vercel function)
+└── cron.js             # Daily cron entry (Vercel Cron)
 src/
-├── index.js              # entry point: wire up config, mongo, scrapers, bot, scheduler
+├── app-builder.js      # wires config, Upstash, scrapers, bot, scheduler
 ├── config.js
 ├── logger.js
 ├── api/
 │   ├── apple-scraper.js
 │   └── google-scraper.js
-├── models/               # plain object factories matching Mongo docs
+├── models/             # plain object factories matching the Mongo schema
-├── repository/           # Mongo collection wrappers (admin / group / cached app)
+├── repository/         # Upstash adapter + per-collection wrappers
 ├── bot/
-│   ├── bot.js            # Telegram polling, command dispatch, sender
+│   ├── bot.js          # command dispatch, sender
-│   └── commands/         # one file per /command
+│   ├── dispatch.js
-├── scheduler/scheduler.js  # daily 7am Vietnam-time check
+│   ├── telegram-api.js
-└── util/                 # table renderer, time helpers
+│   └── commands/       # one file per /command
+├── scheduler/scheduler.js  # 07:00 Asia/Saigon = 00:00 UTC
+└── util/               # table renderer, time helpers
+scripts/
+├── register-webhook.js
+└── check-secret-leaks.js
 ```
 
 ## Differences vs Go / Java
 
 - Group / admin / chat IDs are JS `number`s. Telegram chat IDs fit in safe-int
   range, so this is intentional and matches Telegram's documented limits.
-- Pino logging instead of Java/Go's structured loggers; semantics equivalent.
+- Pino-style structured JSON logging instead of Java/Go's structured loggers.
 - HTTP via Node 20's built-in `fetch` (no extra dependency).
+- Storage is Upstash Redis (REST) instead of MongoDB; key namespace mirrors the
+  original collections, TTL via Redis `EX`.

docker-compose.dev.yml

@@ -1,56 +0,0 @@
-version: '3.8'
-
-services:
-  bot:
-    build: .
-    container_name: js-store-scraper-bot-dev
-    restart: unless-stopped
-    environment:
-      - TELEGRAM_BOT_TOKEN=${TELEGRAM_BOT_TOKEN}
-      - TELEGRAM_BOT_USERNAME=${TELEGRAM_BOT_USERNAME}
-      - MONGODB_CONNECTION_STRING=mongodb://mongodb:27017
-      - MONGO_DATABASE=store_scraper_bot_dev
-      - ENV=DEVELOPMENT
-      - ADMIN_IDS=${ADMIN_IDS}
-      - SOURCE_COMMIT=${SOURCE_COMMIT:-dev}
-    depends_on:
-      - mongodb
-    networks:
-      - bot-network
-    volumes:
-      - ./src:/app/src
-
-  mongodb:
-    image: mongo:7.0
-    container_name: js-store-scraper-mongodb-dev
-    restart: unless-stopped
-    environment:
-      - MONGO_INITDB_DATABASE=store_scraper_bot_dev
-    volumes:
-      - mongodb_data_dev:/data/db
-    networks:
-      - bot-network
-    ports:
-      - "27017:27017"
-
-  mongo-express:
-    image: mongo-express:latest
-    container_name: js-store-mongo-express-dev
-    restart: unless-stopped
-    environment:
-      - ME_CONFIG_MONGODB_URL=mongodb://mongodb:27017
-      - ME_CONFIG_BASICAUTH_USERNAME=admin
-      - ME_CONFIG_BASICAUTH_PASSWORD=admin
-    depends_on:
-      - mongodb
-    networks:
-      - bot-network
-    ports:
-      - "8081:8081"
-
-networks:
-  bot-network:
-    driver: bridge
-
-volumes:
-  mongodb_data_dev:

docker-compose.yml

@@ -1,39 +0,0 @@
-version: '3.8'
-
-services:
-  bot:
-    build: .
-    container_name: js-store-scraper-bot
-    restart: unless-stopped
-    environment:
-      - TELEGRAM_BOT_TOKEN=${TELEGRAM_BOT_TOKEN}
-      - TELEGRAM_BOT_USERNAME=${TELEGRAM_BOT_USERNAME}
-      - MONGODB_CONNECTION_STRING=mongodb://mongodb:27017
-      - MONGO_DATABASE=store_scraper_bot
-      - ENV=${ENV:-PRODUCTION}
-      - ADMIN_IDS=${ADMIN_IDS}
-      - SOURCE_COMMIT=${SOURCE_COMMIT:-unknown}
-    depends_on:
-      - mongodb
-    networks:
-      - bot-network
-
-  mongodb:
-    image: mongo:7.0
-    container_name: js-store-scraper-mongodb
-    restart: unless-stopped
-    environment:
-      - MONGO_INITDB_DATABASE=store_scraper_bot
-    volumes:
-      - mongodb_data:/data/db
-    networks:
-      - bot-network
-    ports:
-      - "27017:27017"
-
-networks:
-  bot-network:
-    driver: bridge
-
-volumes:
-  mongodb_data:

package.json

@@ -13,14 +13,8 @@
     "deploy": "vercel deploy --prod && npm run register",
     "register": "node --env-file=.env.deploy scripts/register-webhook.js",
     "register:dry": "node --env-file=.env.deploy scripts/register-webhook.js --dry-run",
-    "migrate": "node --env-file=.env.deploy scripts/migrate-atlas-to-upstash.js",
-    "migrate:dry": "node --env-file=.env.deploy scripts/migrate-atlas-to-upstash.js --dry-run",
     "lint": "node scripts/check-secret-leaks.js"
   },
-  "devDependencies": {
-    "mongodb": "^6.10.0",
-    "wrangler": "^3.90.0"
-  },
   "license": "Apache-2.0",
   "dependencies": {
     "@upstash/redis": "^1.38.0",

plans/260509-1656-consolidate-vercel-upstash/phase-06-deploy-cutover-and-webhook-reregister.md

@@ -1,7 +1,7 @@
 ---
 phase: 6
 title: "Deploy + cutover + webhook re-register"
-status: pending
+status: completed
 priority: P1
 effort: 45 min
 dependencies: [4, 5]

plans/260509-1656-consolidate-vercel-upstash/phase-07-cleanup-wrangler-and-docs.md

@@ -1,7 +1,7 @@
 ---
 phase: 7
 title: "Cleanup wrangler + docker + docs"
-status: pending
+status: completed
 priority: P2
 effort: 30 min
 dependencies: [6]

plans/260509-1656-consolidate-vercel-upstash/plan.md

@@ -1,7 +1,7 @@
 ---
 title: "Consolidate on Vercel + Upstash Redis"
 description: "Deploy bot to Vercel (no Cloudflare). Migrate live state from legacy java-store-scraper-bot MongoDB Atlas → Upstash Redis. Inline scraper libs (drop store-scraper.vercel.app fetch). Delete Docker artifacts. Single repo, single vendor, free tier."
-status: in-progress
+status: completed
 priority: P1
 effort: 6h
 branch: main
@@ -35,8 +35,8 @@ Storage research: [`reports/researcher-260509-1656-upstash-vs-atlas.md`](../repo
 | 03 | [Inline scraper modules](phase-03-inline-scraper-modules.md) | 45 min | completed |
 | 04 | [HTTP layer (webhook + cron)](phase-04-http-layer-webhook-and-cron.md) | 1h | completed |
 | 05 | [Data migration MongoDB Atlas → Upstash](phase-05-data-migration-atlas-to-upstash.md) | 30 min | completed |
-| 06 | [Deploy + cutover + webhook register](phase-06-deploy-cutover-and-webhook-reregister.md) | 45 min | pending (operator) |
+| 06 | [Deploy + cutover + webhook register](phase-06-deploy-cutover-and-webhook-reregister.md) | 45 min | completed |
-| 07 | [Cleanup wrangler + docker + docs](phase-07-cleanup-wrangler-and-docs.md) | 30 min | pending (post-deploy) |
+| 07 | [Cleanup wrangler + docker + docs](phase-07-cleanup-wrangler-and-docs.md) | 30 min | completed |
 
 ## Key Constraints
 

plans/todo.md

@@ -1,10 +1,10 @@
 # Outstanding Work
 
-Quick index of active direction. CF Workers path superseded — bot now targets Vercel + Upstash.
+Quick index of active direction. CF Workers path superseded — bot runs on Vercel + Upstash.
 
-## Active plan
+## Completed plans
 
-**[260509-1656-consolidate-vercel-upstash](260509-1656-consolidate-vercel-upstash/plan.md)** — Vercel deploy, Atlas → Upstash data migration, Docker + wrangler cleanup. 7 phases, ~6h.
+**[260509-1656-consolidate-vercel-upstash](260509-1656-consolidate-vercel-upstash/plan.md)** — Vercel deploy, Atlas → Upstash data migration, Docker + wrangler cleanup. 7 phases done.
 
 ## Superseded plans (left for history; do not execute)
 

scripts/migrate-atlas-to-upstash.js

@@ -1,158 +0,0 @@
-// One-shot legacy-DB migrator: MongoDB Atlas (java-store-scraper-bot) → Upstash Redis.
-// Direct write — no on-disk JSON intermediate.
-//
-//   common._id="admin"   → SET <prefix>admin <json>
-//   group.find({})       → SET <prefix>group:<_id> <json>     (per group)
-//   apple_app.find({})   → SET <prefix>apple:<_id> <json> EX <ttl>   (only with --include-cache)
-//   google_app.find({})  → SET <prefix>google:<_id> <json> EX <ttl>  (only with --include-cache)
-//
-// KEY_PREFIX defaults to 'store-scraper-bot:' — must match what the bot
-// runtime reads or migrated data is invisible after cutover.
-//
-// Run:    npm run migrate
-// Dry:    npm run migrate:dry
-// Cache:  npm run migrate -- --include-cache
-//
-// Reads .env.deploy for: MONGODB_URI, UPSTASH_REDIS_REST_URL,
-// UPSTASH_REDIS_REST_TOKEN, KEY_PREFIX (optional), APP_CACHE_SECONDS (optional).
-
-import { MongoClient } from 'mongodb';
-import { Redis } from '@upstash/redis';
-
-const MIN_TTL_SECONDS = 60;
-const DEFAULT_KEY_PREFIX = 'store-scraper-bot:';
-const APP_CACHE_SECONDS = Number(process.env.APP_CACHE_SECONDS ?? 600);
-
-function exitWith(message) {
-  console.error(`migrate-atlas-to-upstash: ${message}`);
-  process.exit(1);
-}
-
-function log(line) {
-  console.log(`migrate-atlas-to-upstash: ${line}`);
-}
-
-// Compute remaining TTL in seconds for a cached app entry, given its
-// stored `millis` (= cache write time). Returns null if already expired.
-function remainingTtl(millis, nowMs) {
-  const expiresAt = millis + APP_CACHE_SECONDS * 1000;
-  const remainingSec = Math.floor((expiresAt - nowMs) / 1000);
-  if (remainingSec <= 0) return null;
-  return Math.max(MIN_TTL_SECONDS, remainingSec);
-}
-
-async function main() {
-  const mongoUri = process.env.MONGODB_URI;
-  if (!mongoUri) exitWith('MONGODB_URI not set; check .env.deploy');
-
-  // Accept either Upstash naming convention (vanilla signup) or Vercel
-  // Marketplace integration names — same as the runtime adapter.
-  const upstashUrl = process.env.UPSTASH_REDIS_REST_URL ?? process.env.KV_REST_API_URL;
-  const upstashToken = process.env.UPSTASH_REDIS_REST_TOKEN ?? process.env.KV_REST_API_TOKEN;
-  const dryRun = process.argv.includes('--dry-run');
-  if (!dryRun) {
-    if (!upstashUrl) exitWith('UPSTASH_REDIS_REST_URL (or KV_REST_API_URL) not set; check .env.deploy');
-    if (!upstashToken) exitWith('UPSTASH_REDIS_REST_TOKEN (or KV_REST_API_TOKEN) not set; check .env.deploy');
-  }
-
-  const includeCache = process.argv.includes('--include-cache');
-  const prefix = process.env.KEY_PREFIX ?? DEFAULT_KEY_PREFIX;
-
-  log(`mode:           ${dryRun ? 'DRY RUN (no writes)' : 'LIVE (writes to Upstash)'}`);
-  log(`KEY_PREFIX:     ${prefix}   (must match bot runtime KEY_PREFIX)`);
-  log(`include-cache:  ${includeCache}`);
-
-  const redis = dryRun
-    ? null
-    : new Redis({ url: upstashUrl, token: upstashToken });
-
-  // Long timeout because Atlas free tier auto-pauses idle clusters; first hit
-  // can take 10–30 s to wake up. Migration is one-shot, not perf-critical.
-  const mongo = new MongoClient(mongoUri, {
-    serverSelectionTimeoutMS: 30000,
-    socketTimeoutMS: 30000,
-    appName: 'migrate-atlas-to-upstash',
-  });
-  await mongo.connect();
-  const db = mongo.db();
-
-  const counts = {
-    admin: 0,
-    group: 0,
-    apple: 0,
-    appleSkipped: 0,
-    google: 0,
-    googleSkipped: 0,
-  };
-
-  async function writeKey(logicalKey, value, ttlSeconds = null) {
-    const physical = `${prefix}${logicalKey}`;
-    if (dryRun) {
-      log(`  DRY  SET ${physical}${ttlSeconds != null ? ` EX ${ttlSeconds}` : ''}`);
-      return;
-    }
-    await redis.set(
-      physical,
-      JSON.stringify(value),
-      ttlSeconds != null ? { ex: ttlSeconds } : undefined,
-    );
-  }
-
-  // 1. admin singleton (common._id = "admin")
-  const adminDoc = await db.collection('common').findOne({ _id: 'admin' });
-  if (adminDoc) {
-    await writeKey('admin', adminDoc);
-    counts.admin = 1;
-  } else {
-    log('warning: no admin doc found in common collection');
-  }
-
-  // 2. groups
-  const groupDocs = await db.collection('group').find({}).toArray();
-  for (const doc of groupDocs) {
-    await writeKey(`group:${doc._id}`, doc);
-  }
-  counts.group = groupDocs.length;
-
-  // 3. caches (opt-in)
-  if (includeCache) {
-    const now = Date.now();
-    const appleDocs = await db.collection('apple_app').find({}).toArray();
-    for (const doc of appleDocs) {
-      const ttl = remainingTtl(doc.millis ?? 0, now);
-      if (ttl == null) {
-        counts.appleSkipped++;
-        continue;
-      }
-      await writeKey(`apple:${doc._id}`, doc, ttl);
-      counts.apple++;
-    }
-
-    const googleDocs = await db.collection('google_app').find({}).toArray();
-    for (const doc of googleDocs) {
-      const ttl = remainingTtl(doc.millis ?? 0, now);
-      if (ttl == null) {
-        counts.googleSkipped++;
-        continue;
-      }
-      await writeKey(`google:${doc._id}`, doc, ttl);
-      counts.google++;
-    }
-  }
-
-  await mongo.close();
-
-  log('---');
-  log(`admin:   ${counts.admin}`);
-  log(`groups:  ${counts.group}`);
-  if (includeCache) {
-    log(`apple:   ${counts.apple} (skipped ${counts.appleSkipped} expired)`);
-    log(`google:  ${counts.google} (skipped ${counts.googleSkipped} expired)`);
-  } else {
-    log('caches:  skipped (use --include-cache to migrate them)');
-  }
-  log('');
-  log(dryRun ? 'dry run complete — no Upstash writes performed' : 'migration complete');
-}
-
-main().catch((err) => exitWith(err.stack ?? err.message ?? String(err)));

src/logger.js

@@ -1,5 +1,4 @@
-// Worker-friendly structured logger. Cloudflare Observability indexes JSON
+// Structured JSON logger — one JSON record per call for log aggregators.
-// console output, so we emit one JSON record per call.
 export function createLogger() {
   function log(level, payloadOrMsg, maybeMsg) {
     const isObj = payloadOrMsg !== null && typeof payloadOrMsg === 'object';

src/models/apple-app.js

@@ -1,5 +1,5 @@
 // AppleApp cache entry — Java parity (_id=appId, class="AppleApp").
-// TTL is enforced by Cloudflare KV via expirationTtl, so no isExpired helper.
+// TTL is enforced by Upstash Redis EX, so no isExpired helper.
 export function newAppleApp(appId, response, millis) {
   return { _id: appId, class: 'AppleApp', app: response, millis };
 }

src/models/google-app.js

@@ -1,5 +1,5 @@
 // GoogleApp cache entry — Java parity (_id=appId, class="GoogleApp").
-// TTL is enforced by Cloudflare KV via expirationTtl, so no isExpired helper.
+// TTL is enforced by Upstash Redis EX, so no isExpired helper.
 export function newGoogleApp(appId, response, millis) {
   return { _id: appId, class: 'GoogleApp', app: response, millis };
 }

src/repository/upstash.js

@@ -1,6 +1,6 @@
-// Upstash Redis adapter — replaces the prior Cloudflare KV wrapper.
+// Upstash Redis adapter.
 //
-// Logical key namespace (unchanged from KV layer):
+// Logical key namespace:
 //   admin                       singleton
 //   group:{chatId}              per-group state
 //   apple:{appId}               cached Apple response (with TTL)

wrangler.toml

@@ -1,32 +0,0 @@
-name = "store-scraper-bot"
-main = "src/index.js"
-compatibility_date = "2025-10-01"
-# nodejs_compat_v2 enables node:net + node:tls so the official `mongodb` driver
-# can open a TCP socket to Atlas. v1 vs v2 are alternatives, not additive.
-compatibility_flags = ["nodejs_compat_v2"]
-
-[vars]
-APP_CACHE_SECONDS = "600"
-NUM_DAYS_WARNING_NOT_UPDATED = "30"
-
-# Daily check job. Cloudflare cron is UTC.
-# 0 UTC = 7am Asia/Ho_Chi_Minh (UTC+7).
-[triggers]
-crons = ["0 0 * * *"]
-
-# Workers Observability — captures console.* logs and request metadata in the
-# Cloudflare dashboard. 200k events/day on the Free plan.
-[observability]
-enabled = true
-head_sampling_rate = 1
-
-[observability.logs]
-enabled = true
-invocation_logs = true
-
-# Secrets (set via `wrangler secret put <name>`, NOT in this file):
-#   TELEGRAM_BOT_TOKEN       — bot token from @BotFather
-#   TELEGRAM_BOT_USERNAME    — bot username (without @)
-#   TELEGRAM_WEBHOOK_SECRET  — random ≥32 chars; validates the X-Telegram-Bot-Api-Secret-Token header
-#   MONGODB_URI              — full SRV string, db inferred from URI path
-#   ADMIN_IDS                — comma-separated Telegram user IDs