A public discovery feed for second-hand computers, parts, gaming gear, and tech in Indonesia.

Open RECON

RECON collects public listings from multiple sources and presents them in one clean, searchable feed. Listings are normalized into consistent categories, prices, conditions, locations, and availability, with a direct link back to the original post.

• Browse all discoveries in one place.
• Search and filter by collection, platform, price, status, condition, or location.
• Open a listing to see its images and important details.
• Continue to the original source when something looks interesting.

RECON currently discovers public listings from Facebook Marketplace, Facebook Groups, Instagram, and Reddit.

• Next.js, tRPC, Prisma, and PostgreSQL
• Python collectors and a queued AI normalization pipeline
• Docker Compose, GitHub Actions, and Docker Hub

npm install
npm run db:generate
npm run db:smoke
npm run check
npm run build

Use Docker Compose for local PostgreSQL. Keep real secrets in the ignored .env files.

Requirements: Docker with Compose, a public or local port for the web app, and outbound internet access for the scrapers.

Copy-Item .env.example .env
docker compose up -d --build postgres web
docker compose exec web npx prisma migrate deploy

Before starting, edit .env:

• Replace POSTGRES_PASSWORD and keep the same database name, user, and password in DATABASE_URL.
• Set NVIDIA_API_KEY to enable AI normalization.
• Set all five R2_* values to enable the media worker and copy Instagram, Facebook Group, and Reddit images to Cloudflare R2.
• Keep SCRAPER_EGRESS_MODE=direct unless you intentionally configure and allow a proxy or VPN.

Open http://localhost:3000. To run the scheduled scraper pipeline:

docker compose up -d --build scraper-scheduler scraper-ai-manager
docker compose ps
docker compose logs -f scraper-scheduler scraper-ai-manager

After configuring R2, also start scraper-media-worker. The scheduler collects listings, the AI manager normalizes and writes them to PostgreSQL, and the optional media worker caches Instagram, Facebook Group, and Reddit images. Edit accounts, source enablement, limits, and schedules in scraper/config/sources.toml; edit Marketplace searches in scraper/facebook/source_targets.json and Group targets in scraper/facebook_groups/source_targets.json. Never put API keys, cookies, or passwords in those tracked files.

RECON is available under the MIT License.