# llmmcp

Feeds your agent current model and API details from OpenAI, Anthropic and Google so it stops inventing versions and parameters.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed · ★ 6
- Author: zero-abd
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/zero-abd/llmmcp
- Updated: 2026-02-18
- HTML: https://affpapa.org/en/nevibecode/mcp/llmmcp

## What it does

\`llmmcp` is an MCP server that hands your coding agent official provider documentation instead of whatever it half-remembers from training. It exposes two tools: `search_docs` for looking things up in the indexed docs, and `list_providers` for the current provider and model lineup.

Under the hood a scraper pulls markdown and text from the official sites once a week. Chunks land in Pinecone with integrated embeddings, and a Cloudflare Worker handles embedding and retrieval, caching popular queries in Workers KV. A small CLI turns MCP calls into requests to that Worker. Nothing runs locally: `npx -y llmmcp@latest` starts it, and Cursor takes it as a command under Settings > Models > MCP Servers while Claude Desktop wants an entry in `claude_desktop_config.json`.

Queries leave your machine for a hosted backend, so if prompts or internal docs must stay local, sort out self-hosting first. The repo is open source, though the deployment guide is not written yet — the README points at the logs for the current setup.

## Install

### Claude Code (native platform)

**Install**
- The MCP registry has no launch command for it: see the repository README

## When to use it
- check tool use parameters for a specific model
- confirm context window sizes and rate limits
- list a provider's current models without hardcoding them
- steer an agent toward a current library pattern instead of a deprecated one

## Example prompts
- Look up the tool use parameters for Gemini 1.5 Pro in the docs
- Which Anthropic models are current right now?
- Check the official docs for how to set rate limits

## Similar
- [REA](https://affpapa.org/en/nevibecode/mcp/rea.md) — MCP server · ★ 386989
- [GitHub](https://affpapa.org/en/nevibecode/mcp/github.md) — MCP server · ★ 33918
- [Netdata](https://affpapa.org/en/nevibecode/mcp/netdata.md) — MCP server · ★ 81115
- [browser-use](https://affpapa.org/en/nevibecode/mcp/browser-use.md) — MCP server · ★ 119160
- [Codebase Memory](https://affpapa.org/en/nevibecode/mcp/codebase-memory.md) — MCP server · ★ 47392
- [mcp-server](https://affpapa.org/en/nevibecode/mcp/mcp-server.md) — MCP server · ★ 43872

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md