Terobyte/librarian-cli
by Various
📚 Any ebook → clean token-counted Markdown chapters. Let Claude read your bookshelf over MCP — and catch it misquoting your books. Deterministic RAG: no embeddings, no network, no
MCP
Terobyte/librarian-cli
Added 8 Sept 2026
Overview
A Python CLI that converts ebooks into clean, token-counted Markdown chapters. It lets Claude read books over MCP and uses deterministic RAG with no embeddings, network, or API keys. Designed to catch misquotations by Claude.
Best for
Best for
Developers who want a deterministic, offline way to feed ebooks to Claude via MCP.
Use cases
- Convert ebooks to Markdown chapters for LLM ingestion
- Expose a bookshelf to Claude via MCP
- Run deterministic retrieval without embeddings or external services
How to use
Install
claude mcp add librarian --env LIB_HOME=$HOME/books -- uvx librarian-cliTools exposed
not_found
Clients listed in this record (not verified here)
Claude Desktop, Claude Code
Example client config
{
"mcpServers": {
"librarian": {
"command": "uvx",
"args": [
"librarian-cli"
],
"env": {
"LIB_HOME": "/path/to/books"
}
}
}
}Notes
2 stars on GitHub. Last updated 2026-09-01. Licensed MIT.
Pros
- No external dependencies for retrieval (no embeddings, network, API keys)
- Token-counted output helps manage context limits
- Open source Python CLI
Cons
- Very low community adoption (2 GitHub stars)
- Focused on ebooks, not arbitrary document formats
- Requires MCP setup for Claude integration
Open-source & AI alternatives
Swap-in tools that solve the same job. Weigh the trade-offs before you commit.
Get the free Developer’s Field Guide
A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.
Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.
