AI-powered company enrichment tool. Give it any domain and get back a detailed company profile with firmographic, technographic, and hiring data — exported as clean Markdown files.
Built with Claude (via Vercel AI SDK), runs entirely on your machine.
Point it at a company domain and it will:
- Scrape the company website for basic info, social links, and tech stack detection
- Pull GitHub data — repos, stars, languages, activity
- Scrape job boards (Greenhouse, Lever, Ashby) for open positions and hiring signals
- Detect mobile apps from App Store / Play Store links
- Estimate engineering headcount from multiple signals
- Use Claude AI to synthesize everything into structured data with competitive insights
The output is a well-formatted Markdown report you can read, share, or feed into other tools.
# 1. Install dependencies
npm install
# 2. Set up your API key
cp .env.example .env
# Edit .env and add your Anthropic API key
# Get one at https://console.anthropic.com
# 3. Build the project
npm run build
# 4. Run the web UI
npm run dev
# Open http://localhost:3000📖 New to this tool? Read the full guide: GETTING_STARTED.md
npm run devOpens a local web interface at http://localhost:3000. Paste one or more domains (one per line), click Enrich, and get results instantly. The UI has a Mac-native feel with live Markdown preview.
# Single domain
npm run enrich -- stripe.com
# Multiple domains
npm run enrich -- stripe.com anthropic.com vercel.com
# From a file (one domain per line)
npm run enrich -- --file domains.txtResults are saved as .md files in the ./output/ directory.
Want a native macOS app instead of running in your browser?
# Install Electron dependencies (first time only)
npm install
# Build the .app bundle
npm run electron:build
# Or build a universal binary (Intel + Apple Silicon)
npm run electron:distThe app will be in release/mac/ - drag it to your Applications folder.
npm run electron:devThis builds the project and launches the Electron app with DevTools enabled.
The desktop app:
- Runs the same Express server as the web version
- Opens it in a native macOS window
- Uses your existing
.envconfiguration - Exports files to the same
output/directory
Note: The web server (npm run dev) still works exactly as before - Electron is purely optional.
📖 Learn more: ELECTRON.md for detailed Electron documentation.
Copy .env.example to .env and set:
| Variable | Required | Description |
|---|---|---|
ANTHROPIC_API_KEY |
Yes | Your Anthropic API key (get one here) |
GITHUB_TOKEN |
No | GitHub personal access token — increases API rate limits from 60 to 5,000 requests/hour |
PORT |
No | Server port (default: 3000) |
Each enrichment report includes (when available):
- Company overview — name, description, founded year, employee count, HQ location
- Funding history — rounds, amounts, investors, valuations
- Leadership — CEO, founders, recent leadership changes
- Tech stack — frontend/backend frameworks, databases, cloud providers, analytics, payments, auth, CDN
- Hiring data — open positions by department, top skills being hired for, individual job listings
- GitHub activity — public repos, stars, forks, programming languages, top repositories
- Mobile apps — iOS and Android app detection
- AI insights — growth stage, competitive landscape, key differentiators, recent news
src/
index.ts # Express server + web UI
scripts/
enrich.ts # CLI entry point
lib/
enrichment-engine.ts # Core orchestrator — coordinates all data sources + AI
tech-detector.ts # Detects tech stack from HTML
job-scraper.ts # Scrapes Greenhouse, Lever, Ashby job boards
github-fetcher.ts # GitHub API integration
linkedin-headcount.ts # Engineering headcount estimation
mobile-app-detector.ts # iOS/Android app detection from HTML
markdown-exporter.ts # Formats enrichment data as Markdown
types/
enrichment.ts # TypeScript interfaces for all data structures
- Runtime: Node.js + TypeScript
- AI: Claude via Vercel AI SDK + @ai-sdk/anthropic
- Web scraping: Axios + Cheerio
- Server: Express (serves both the API and the web UI)
- Validation: Zod (structured AI output)
MIT