Ferret Documentation

The universal API that connects AI agents to the web in real-time. 333 parsers, 363 routes, 121 MCP tools, 17 business endpoints — one self-hosted API.

Quick Start →

Get started in 3 minutes. Build, query, deploy.

API Reference →

Search, Extract, Crawl, Map, Research endpoints.

MCP Server →

Cursor, Claude Desktop, Claude Code, OpenAI integration.

SDKs →

Python, JavaScript, LangChain integrations.

Key Features

✓ 333 web parsers — search, e-commerce, finance, news, SEO, social, and 49 France-specific sources
✓ 121 MCP tools — native Model Context Protocol support
✓ 17 business endpoints — cross-cutting analyses with LLM-ready markdown prompts
✓ Rust architecture — cold start under 100ms, 46MB RAM
✓ Self-hosted — MIT license, your infrastructure, your rules
✓ Tavily-compatible — drop-in replacement for Extract and Research APIs

Install & Run

git clone https://github.com/duan78/ferret.git
cd ferret
cargo build --release
./target/release/ferret serve    # REST API on :9093
./target/release/ferret mcp      # MCP server on :9094