# NotebookLM MCP

MCP server for NotebookLM - Let your AI agents (Claude Code, Codex) research documentation directly with grounded, citation-backed answers from Gemini. Persistent auth, library management, cross-client sharing. Zero hallucinations, just your knowledge base.

Repository: https://github.com/PleasePrompto/notebooklm-mcp
Canonical: https://ross.abutalabs.com/products/notebooklm-mcp
Language: TypeScript
License: MIT
License Family: permissive
Last push: 2026-05-01T05:51:20+00:00

## Health v2 (maintenance only)
Score: 65/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 80, release rhythm 69, longevity 22
- inputs: {"age_days": 321, "days_push": 124, "days_rel": 124, "gap_med": 35, "n_releases_24m": 4}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3350, forks 476 (observed 2026-08-28T04:07:57.635164+00:00)

## What it is
An MCP server and Claude Code skill that lets AI coding agents chat directly with Google NotebookLM notebooks, returning source-grounded, citation-backed answers from Gemini based only on the user's uploaded documents. It provides browser automation, notebook library management, persistent authentication, source ingestion, and audio overview generation over stdio and Streamable-HTTP transports.

## Use cases
- query my uploaded documents from claude code with citation-backed answers
- connect an mcp client to google notebooklm notebooks
- reduce hallucinations by grounding coding agent answers in my own docs
- ingest sources into notebooklm and generate audio overviews programmatically
- manage a library of notebooklm notebooks with persistent login
- ask questions about my project documentation without copy-pasting into a browser

## When to choose
- you already use Claude Code or another MCP client and want NotebookLM as a grounded knowledge source
- you want answers restricted to your uploaded documents with citations instead of keyword search over local files
- you prefer NotebookLM's Gemini-based synthesis over building and maintaining a local RAG pipeline
- you need to automate notebook chat, source ingestion, or audio overview creation

## When to avoid
- you need a fully offline or self-hosted RAG solution without a Google account
- you want to use the skill inside the Claude Code web UI, which lacks the network access browser automation requires
- you are not comfortable with browser automation against the NotebookLM web interface, which may break if Google changes it
- you need a general-purpose web search or public-internet question answering rather than answers from your own corpus

## Facets
- artifact type: service
- maturity: active
- function: mcp, rag, agent-framework, browser-extension
- domain: large-language-models, developer-tools, artificial-intelligence
- platform: python, cli
- tags: notebooklm, google-gemini, claude-code, browser-automation, citation-backed-answers, knowledge-base, source-grounded, persistent-authentication, audio-overviews, stdio-transport, streamable-http, retrieval-augmented-generation, nodejs, docker

## Member repositories
- PleasePrompto/notebooklm-mcp (main) score 65
- PleasePrompto/notebooklm-skill (plugin) score 48

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:57.635164+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:35:27.804904+00:00, confidence not recorded.
  - readme: https://github.com/PleasePrompto/notebooklm-mcp (fetched 2026-08-28T04:07:57.635164+00:00, sha 68ba4c4cb2b3)
  - registry_npm: https://registry.npmjs.org/notebooklm-mcp (fetched 2026-08-29T08:31:48.386446+00:00, sha f2a964820646)
- Data as of 2026-08-30T08:39:29.467469+00:00.