@usecrow/envoy

CLI that reads your backend codebase, generates an MCP server, and deploys it to Crow — so AI agents can call your API.

Quick Start

export CROW_API_KEY=ck_...
npx @usecrow/envoy

Run from the root directory that contains your project folders (backend, frontend, etc.). The envoy will ask which directories are relevant before it starts.

Commands

| Command | Description | |---------|-------------| | npx @usecrow/envoy | Generate an MCP server from your codebase (default) | | npx @usecrow/envoy fix | Fix/iterate on an existing MCP server | | npx @usecrow/envoy deploy | Deploy the latest saved version |

Options

| Flag | Description | |------|-------------| | --api-key <key> | Crow API key (alternative to CROW_API_KEY env var) | | --ci | CI mode — structured JSON output on stdout | | --dirs <a,b,c> | Comma-separated directories to scan (CI mode, skips interactive prompt) | | --help, -h | Show help | | --version, -v | Show version |

Environment Variables

| Variable | Required | Description | |----------|----------|-------------| | CROW_API_KEY | Yes | Your Crow API key | | CROW_API_URL | No | Override API URL (for development) |

Get your API key at app.usecrow.com → Tools.

How It Works

1. Discover — Scans your project directories for routes, schemas, and API structure
2. Generate — An AI agent analyzes your codebase and builds a complete MCP server
3. Save — Pushes the generated server to Crow with versioning
4. Deploy — Makes the MCP server live so AI agents can call your API

License

MIT