rtfmbro-mcp

Name: rtfmbro-mcp
Author: marckrenn

@marckrenn

About rtfmbro-mcp

rtfmbro provides always-up-to-date, version-specific package documentation as context for coding agents. An alternative to context7

#agent #documentation #docs #mcp #claude #agentic-ai #mcp-server #context7

Basic information

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "rtfmbro": {
      "type": "http",
      "url": "https://rtfmbro.smolosoft.dev/mcp/"
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is rtfmbro-mcp?

rtfmbro-mcp is a Model Context Protocol (MCP) server that fetches version-specific package documentation from GitHub repositories in real time. It integrates with AI coding assistants to provide always‑up‑to‑date documentation for Python (PyPI) and Node.js (npm) packages, with Swift (SPM) in alpha and a GitHub fallback. It is an alternative to context7.

How to use rtfmbro-mcp?

Add the server to your MCP configuration using the HTTP URL https://rtfmbro.smolosoft.dev/mcp/. Optionally copy the provided meta-instruction file for your agent—e.g., .github/copilot-instructions.md for GitHub Copilot or CLAUDE.md for Claude Code—to your project root. Once configured, AI agents can call four tools: get_readme, get_documentation_tree, read_files, and search_github_repositories.

Key features of rtfmbro-mcp

Fetches docs from the exact git tag matching your package version
Supports Python (PyPI) and Node.js (npm) fully
SHA‑based caching with automatic invalidation on new commits
Four primary MCP tools for reading docs and searching repos
Agent integration via project‑root instruction files (Copilot, Claude, Cursor)
Zero setup for developers—just add the server config and start fetching

Use cases of rtfmbro-mcp

Getting version‑precise documentation for legacy projects pinned to an old release
Providing AI agents with fresh docs for brand‑new packages not in training data
Replacing inefficient manual browsing of node_modules or local source code
Speeding up code generation by supplying high‑level docs as context