rtfm
Active·★ 25·MIT·Updated 2026-10-18
RTFM is a local, open-source retrieval layer that indexes everything in a project (code, docs, PDFs, etc.) so that AI agents can find the right context instantly without hallucinations or cloud costs. It integrates with Claude Code and other MCP clients, supports multiple formats, and provides fast search with progressive disclosure.
#ai-agents#claude#claude-code#code-search#context-engineering#developer-tools#embeddings#fts
01
Features
01Full-text, semantic, and hybrid search with progressive disclosure
02Multi-format indexing with 22 built-in parsers (code, docs, PDFs, etc.)
03Local, free, and open-source (MIT license)
04Knowledge graph with wikilinks and import edges, plus version history
05Seamless MCP integration (native Claude Code plugin and manual install)
02
Compatibility
Python
Python 3.10+
Verified via docs
Claude Code
Claude Code Plugin
Verified via docs
Linux
Linux
Verified via docs
macOS
macOS
Verified via docs
Windows
Windows (incl. WSL)
Verified via docs
03
Quick start
1
$ pip install rtfm-ai
2
$ cd your-project
3
$ rtfm init
04
Use cases
↳Legal and regulatory document retrieval with support for specialized formats
↳Research projects combining code, papers, and datasets
↳Any project mixing code with domain-specific documents (FinTech, HealthTech, etc.)
05
Alternatives
TrueMemory★ 381
A living memory system that ingests long-horizon data to infer insights, enabling more decisive action, all while running on a single SQLite file locally.
knowledge-rag★ 300
Drop docs, search instantly from Claude Code — 12 MCP tools, 20 format parsers, hybrid search + reranking. Zero servers, zero API keys, 100% local.
ClawMem★ 211
On-device context engine and memory for AI agents. Claude Code, Hermes and OpenClaw. Hooks + MCP server + hybrid RAG search.
claude-historian-mcp★ 179
📜 An MCP server for conversation history search and retrieval in Claude Code
Comments
Log in to leave a comment
No comments yet. Be the first!