Rtfm MCP Server

Index everything in your project — code, docs, PDFs, legal texts, research, data — and your agent finds the right context instantly. No

Local serverstdioPython

What is the Rtfm MCP server?

Index everything in your project — code, docs, PDFs, legal texts, research, data — and your agent finds the right context instantly. No hallucinations. No cloud. No API costs. Exposed over MCP by the rtfm mcp server, that capability becomes something an assistant can invoke while it works, not something you go and do afterwards.

Its toolset

Everything the assistant can do here goes through one of these:

  • rtfm_search — Search the index (FTS, semantic, or hybrid)
  • rtfm_context — Get relevant context for a subject (metadata-only)
  • rtfm_expand — Show all chunks of a source with full content
  • rtfm_discover — Fast project structure scan (~1s, no indexing needed)
  • rtfm_books — List indexed documents
  • rtfm_stats — Library statistics
  • rtfm_sync — Sync a directory (incremental)
  • rtfm_ingest — Ingest a single file
  • rtfm_tags — List all tags
  • rtfm_tag_chunks — Add tags to specific chunks
  • rtfm_remove — Remove a file from the index
  • rtfm_graph — Show dependency graph for a source (imports, links)

Adding it to your client

rtfm-ai on PyPI is all you need. Most clients run it directly, so configuration is a few lines and a restart.

When to reach for it

Plenty of file and storage access servers cover similar ground. The differences that matter in practice are scope of access and how much setup stands between you and a working tool call. Rtfm's toolset — rtfm_search, rtfm_context, rtfm_expand and 11 more — is a fair guide to whether it matches your workflow. It is maintained by roomi-fields; worth a glance at recent repository activity before you build anything load-bearing on it.

SyncDev reviews every entry in this directory against the project's own documentation before publishing, and revisits them as servers change.

Caveats

  • It runs with your machine's permissions. That is convenient and also the reason to think about what you point it at before you approve a tool call.
  • With 14 tools registered it takes up a noticeable share of the context window; turn it off in projects that never touch Rtfm.
  • MCP clients confirm each tool call by default. Leave that on until you have watched what the rtfm mcp server does with a few real requests.

Available tools

ToolWhat it does
rtfm_searchSearch the index (FTS, semantic, or hybrid)
rtfm_contextGet relevant context for a subject (metadata-only)
rtfm_expandShow all chunks of a source with full content
rtfm_discoverFast project structure scan (~1s, no indexing needed)
rtfm_booksList indexed documents
rtfm_statsLibrary statistics
rtfm_syncSync a directory (incremental)
rtfm_ingestIngest a single file
rtfm_tagsList all tags
rtfm_tag_chunksAdd tags to specific chunks
rtfm_removeRemove a file from the index
rtfm_graphShow dependency graph for a source (imports, links)
rtfm_historyFile version history and memory snapshots
IntegrationThe Integration tool exposed by this server.

How to install the Rtfm MCP server

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

Add to claude_desktop_config.json, then restart Claude Desktop.

Example prompts to try

  • Use Rtfm to rtfm search.
  • Use Rtfm to rtfm context.
  • Use Rtfm to rtfm expand.

Frequently asked questions

It connects Rtfm to MCP-compatible AI assistants such as Claude and Cursor, exposing 14 tools (rtfm_search, rtfm_context, rtfm_expand, and more) that the assistant can call on your behalf. Instead of copying data back and forth by hand, the assistant works with Rtfm directly.