Ratel

Context engineering for AI agents. ~80% fewer tokens. Fix tool overload. Skills and memory with in-process BM25 and semantic retrieval. Progressive Disclosure.…

LLM Mart 1 views 220 listing impressions

Ratel is a self-improvement layer for AI agents. Ratel finds problems, fixes them, and runs evals on your agent. Ratel can import traces from observability tools like OpenTelemetry, Langfuse, LangSmith, and PostHog.

Ratel allows replaying past sessions with a fix to evaluate impact on accuracy, cost, and latency. Approved fixes can be shipped live without redeploy via the Ratel SDK.

Summary drafted from the project's own website. Every sentence is backed by text on that page and was reviewed before publishing.

Related servers

Fetch Official

Fetch a web page and hand the agent readable text instead of raw HTML.

stdio
mcp-server-fetch 33 views
Filesystem Official

Give an agent scoped read and write access to directories you choose.

stdio
@modelcontextprotocol/server-filesystem 29 views
Git Official

Read repository history, diffs and branches as tools an agent can call.

stdio
mcp-server-git 27 views

Read-only MCP server for the OrchestKit docs: full-text search + Markdown fetch. No auth.

24 views