RTFM FAQs

Question 1

What is RTFM and how does it help AI agents?

Accepted Answer

RTFM (Retrieve The Forgotten Memory) is an open retrieval layer that indexes your entire project—code, documentation, legal, research, and data—to provide AI agents with precise, on-demand context. This prevents AI hallucinations and ensures agents work with accurate, comprehensive information.

Question 2

What types of files can RTFM index and search?

Accepted Answer

RTFM boasts 10+ built-in parsers for common formats like Markdown, Python, LaTeX, YAML, JSON, Shell scripts, HTML, XML, and PDF. Its flexible architecture also allows developers to easily create custom parsers for any unique file type, extending its comprehensive indexing capabilities.

Question 3

How does RTFM integrate with existing AI agent workflows?

Accepted Answer

RTFM integrates natively as an MCP (Model Context Protocol) server, making it compatible with clients like Claude Code, Cursor, and Codex. It provides a CLI and Python API for direct interaction, along with auto-sync hooks to keep your indexed knowledge base fresh without manual intervention.

Question 4

How does RTFM differ from other code indexing tools?

Accepted Answer

While many tools index code, RTFM indexes *everything*. It goes beyond code to include documentation, legal documents, research papers (LaTeX, PDF), and diverse data formats. It also offers highly extensible parsing capabilities, allowing you to add support for any custom format in minimal lines of Python.

Question 5

Is RTFM open source and what are its dependencies?

Accepted Answer

Yes, RTFM is open source under the MIT license, making it free to use, fork, and extend. It uses a built-in SQLite database, minimizing external dependencies and avoiding the need for expensive cloud services or enterprise servers, offering a lightweight and self-contained solution.

RTFM

RTFM

Key Features

Use Cases

Key Features

Use Cases