Skip to content
MCP server

RTFM

Index code, docs, PDFs and data in a local SQLite file and serve context to agents

Runs locallyStatus and history
Pricing
Open source
Platforms
CLI, Linux, macOS
Runs
Locally
Version
1.0.0
Listing updated
3 days ago

About RTFM

RTFM is an open-source MCP server for knowledge and memory. It runs locally on your machine and is started by your AI client.

RTFM exists because its author watched Claude Code grep the same folders in loops while writing a French tax article from about 50 pages of regulation. The fix is retrieval: one query returns ranked pointers, about three results and 300 tokens, and the agent calls rtfm_expand(source, target_section) to read more.

Everything lives in .rtfm/library.db, a single SQLite file. Search defaults to FTS5, with optional local embeddings through FastEmbed and ONNX. It has 24 built-in parsers covering Python AST, Markdown headers, XML legal text, LaTeX, CSV and XLSX, Jupyter notebooks and SQLite, and you can add formats with about 50 lines of Python or 30 lines of YAML.

On the author's own task the site reports a cost drop from $22.61 to $11.14. That figure is the project's benchmark, not an independent test.

Install

Run in a terminal

claude mcp add rtfm -- uvx rtfm-ai

Add to claude_desktop_config.json

{
  "mcpServers": {
    "rtfm": {
      "command": "uvx",
      "args": [
        "rtfm-ai"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "rtfm": {
      "command": "uvx",
      "args": [
        "rtfm-ai"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "rtfm": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "rtfm-ai"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "rtfm": {
      "command": "uvx",
      "args": [
        "rtfm-ai"
      ]
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "rtfm": {
      "command": "uvx",
      "args": [
        "rtfm-ai"
      ]
    }
  }
}

Tools this server provides (1)

Taken from the project documentation; the running server may differ.

rtfm_expand
Read more of a search hit by source and target section.

What it does

Single SQLite index
One local file holds the whole knowledge base.
24 parsers
Code, Markdown, legal XML, LaTeX, tables, notebooks and databases.
Expandable results
Short ranked hits, then rtfm_expand for a section.
Local only
No cloud, API keys or telemetry.
Declarative mappings
Map JSON schemas to chunks with YAML.
Obsidian vault mode
Search large note vaults.

Good for

  • Find the right spec section before editing code
  • Search PDFs and regulations alongside source
  • Cut down grep loops in a large repo
  • Index an Obsidian vault for an agent

Pricing

Open source

MIT licence; free and local

Prices change. Check rtfm.roomi-fields.com before you buy.

RTFM FAQ

Does it need a cloud service?

No. The docs say it runs locally with no accounts, API keys or quotas.

Is it a vector database?

No. Default search is FTS5; embeddings are optional and local.

How do I start?

Install the Claude Code plugin or pip install rtfm-ai, then run rtfm init in your project.

What licence?

MIT.

Is RTFM free?

Yes. RTFM is open source, so you can use and self-host it for free.

How do I add RTFM to Claude Code?

Run this command in your terminal: claude mcp add rtfm -- uvx rtfm-ai

What are the best alternatives to RTFM?

The closest alternatives to RTFM are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to RTFM

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools