What is Ratel?
Ratel is an open-source context gateway for AI agents: it keeps your tool and skill catalog out of the prompt and retrieves only what each turn needs.
Keep reading to learn more, or get started right away.
Why Ratel?
- Lean context: keeps every tool schema out of the prompt with progressive disclosure, and cuts context with benchmark-backed savings.
- Zero setup: in-process BM25 keyword search: no vector database, no API key, no service; semantic & hybrid ranking when you opt in.
- One catalog: tools, MCP servers, and skills, all searchable and runnable through the same three capability tools.
- Two SDKs: TypeScript and Python over one Rust engine, with Vercel AI SDK and Pydantic AI integrations.
- Observable: the whole search-and-invoke funnel as local trace events or OpenTelemetry spans.
- Open: Apache-2.0 engine, MIT everything else, measured in the open by ratel-bench.
Get started
Follow the Quick start (the fastest path lets your coding agent integrate Ratel for you), or install the SDK and jump straight to the TypeScript or Python quickstart.
What the skills do, how they chain, and how observability fits is on Skillset.
What's new tracks releases, and the FAQ answers common questions.
Using Claude Code or Codex? Ratel Local puts the same capability contract in front of your existing MCP servers.
Features
Everything Ratel does, one page per use case; three are coming soon.
Optimize agent capabilities
Keep the model's tool surface minimal and retrieval sharp: progressive disclosure, BM25 keyword search, and opt-in semantic and hybrid ranking.
Observe your agent
The whole search-and-invoke funnel as local trace events, or exported as OpenTelemetry spans.
Self-improving agent (coming soon)
Ranking and retrieval that learn from your agent's real usage: which capabilities actually get invoked.
Managing chat history (coming soon)
Keep long-running conversations lean without losing what matters.
Agent memory (coming soon)
Long-lived memory your agent can store and retrieve like any other capability.
How it works
Curious? Read Architecture for the turn-by-turn flow, Tools, MCP & Skills for what the catalog holds, and Optimize agent capabilities for exactly what changes in the model's context.
Support and issues
For bugs and questions, open an issue on github.com/ratel-ai/ratel.
Community
Join our brand-new Discord to ask questions and follow development.
