
RTFM
RTFM is a local retrieval layer that indexes code, docs, PDFs, and structured content into a single SQLite database. It exposes MCP tools for search (full-text, semantic, hybrid) and progressive expansion, so agents query indexed metadata first and load only relevant sections instead of grepping blindly through thousands of files. Ships as a Claude Code plugin with auto-indexing on every prompt, or install via pip for manual use with other MCP clients. Optional embeddings and PDF parsing install into an isolated venv. Built for projects where context spans multiple formats (legal XML, research papers, architecture decisions, regulatory text) and you need the agent to find the right paragraph without dumping entire file trees into context.
- GitHub stars
- 27
- tools
- 14
- transport
- STDIO
- last updated
- Oct 30, 2026
Tools
14 toolsPublic tool metadata for what this MCP can expose to an agent.
rtfm_searchSearch the index (FTS, semantic, or hybrid)
Search the index (FTS, semantic, or hybrid)
No parameter schema in public metadata yet.
rtfm_contextGet relevant context for a subject (metadata-only)
Get relevant context for a subject (metadata-only)
No parameter schema in public metadata yet.
rtfm_expandShow all chunks of a source with full content
Show all chunks of a source with full content
No parameter schema in public metadata yet.
rtfm_discoverFast project structure scan (~1s, no indexing needed)
Fast project structure scan (~1s, no indexing needed)
No parameter schema in public metadata yet.
rtfm_booksList indexed documents
List indexed documents
No parameter schema in public metadata yet.
rtfm_statsLibrary statistics
Library statistics
No parameter schema in public metadata yet.
rtfm_syncSync a directory (incremental)
Sync a directory (incremental)
No parameter schema in public metadata yet.
rtfm_ingestIngest a single file
Ingest a single file
No parameter schema in public metadata yet.
rtfm_tagsList all tags
List all tags
No parameter schema in public metadata yet.
rtfm_tag_chunksAdd tags to specific chunks
Add tags to specific chunks
No parameter schema in public metadata yet.
rtfm_removeRemove a file from the index
Remove a file from the index
No parameter schema in public metadata yet.
rtfm_graphShow dependency graph for a source (imports, links)
Show dependency graph for a source (imports, links)
No parameter schema in public metadata yet.
rtfm_historyFile version history and memory snapshots
File version history and memory snapshots
No parameter schema in public metadata yet.
rtfm_coverageWhat the index holds against what is on disk
What the index holds against what is on disk
No parameter schema in public metadata yet.
Sponsors
Readme
RTFM
Retrieve The Forgotten Memory
The open retrieval layer your AI agent was missing
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.
Fewer calls per search: one query returns ranked hits across code, docs and PDFs — a few hundred tokens of pointers — instead of a chain of grep, glob and whole-file reads.
Free · Local · Open Source · MIT
RTFM vs vanilla Claude Code — same task, same model, who pays the bill?
Details
- Repository
- github.com/roomi-fields/rtfm
- Website
- rtfm.roomi-fields.com
- Categories
- Databases, Search & Web Crawling, Data & Analytics
- Package
rtfm-ai- Transport
- STDIO
- Registry
- MCP registry, active