agentleFS
Sign inStart free
jakubs2623jakubs2623/notebooklm-skill/AGENTS.md

notebooklm-skill

AGENTS.md7 starsChanged 8 months agoMITView on GitHub

NotebookLM research automation — CLI, MCP server, and Claude Code Skill. This project bridges Google NotebookLM's research capabilities with AI content generation. Feed it URLs, PDFs, or trending topics — it creates NotebookLM notebooks, runs deep research, and produces structured output: articles, social posts, podcasts, videos, slides, and more. Built on notebooklm-py v0.3.4 — pure async Python. NotebookLM uses browser-based Google login (no API keys needed): Session stored at ~/.notebooklm/storage_state.json. Lasts weeks. Three global commands are available after pip install…

  • Installs packages
  • Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.

    Of the AGENTS.md files we read, the 31,774 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.

  • In this repositoryOne other file it gives its agents.

The file

We split this AGENTS.md into the rules it gives an agent

2 rules. About 0 are common, 0 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    All CLI commands output JSON to stdout.

    line 93
  2. Very specific:

    Use --output to save artifacts to files.

    line 94

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
AGENTS.md

notebooklm-skill

NotebookLM research automation — CLI, MCP server, and Claude Code Skill.

Overview

This project bridges Google NotebookLM's research capabilities with AI content generation. Feed it URLs, PDFs, or trending topics — it creates NotebookLM notebooks, runs deep research, and produces structured output: articles, social posts, podcasts, videos, slides, and more.

Built on notebooklm-py v0.3.4 — pure async Python.

Authentication

NotebookLM uses browser-based Google login (no API keys needed):

python3 -m notebooklm login          # One-time browser auth
python scripts/auth_helper.py verify  # Verify session

Session stored at ~/.notebooklm/storage_state.json. Lasts weeks.

CLI Commands

Three global commands are available after pip install .:

notebooklm-skill — Core Operations

notebooklm-skill create --title "Research" --sources https://example.com
notebooklm-skill list
notebooklm-skill ask --notebook "Research" --query "Key findings?"
notebooklm-skill generate audio --notebook "Research" --language en
notebooklm-skill download audio --notebook "Research" --output podcast.m4a
notebooklm-skill delete --notebook "Research"

notebooklm-pipeline — Workflow Orchestration

notebooklm-pipeline research-to-article --sources url1 url2 --title "Topic"
notebooklm-pipeline research-to-social --sources url1 --platform threads
notebooklm-pipeline trend-to-content --geo TW --count 5 --platform threads
notebooklm-pipeline batch-digest --rss https://example.com/feed.xml
notebooklm-pipeline generate-all --sources url1 --title "Research" --output-dir ./output

notebooklm-mcp — MCP Server

notebooklm-mcp            # stdio mode (Claude Code, Cursor)
notebooklm-mcp --http     # HTTP mode on port 8765

MCP Tools (13)

ToolDescription
nlm_create_notebookCreate notebook with sources
nlm_listList all notebooks
nlm_deleteDelete a notebook
nlm_add_sourceAdd source to existing notebook
nlm_askAsk question (returns answer + citations)
nlm_summarizeGet notebook summary
nlm_generateGenerate artifact (9 types, infographic excluded)
nlm_downloadDownload generated artifact
nlm_list_sourcesList sources in notebook
nlm_list_artifactsList generated artifacts
nlm_researchDeep web research
nlm_research_pipelineFull research pipeline
nlm_trend_researchTrend-to-research pipeline

Artifact Types (9 downloadable)

audio, video, slides, report, quiz, flashcards, mind-map, data-table, study-guide

⚠️ infographic: generation works but download is unreliable. Use slides instead.

Project Structure

scripts/                  CLI wrappers (notebooklm_client.py, pipeline.py)
mcp_server/               FastMCP server (server.py, tools.py)
SKILL.md                  Claude Code Skill definition
docs/                     Setup guides (EN + zh-TW)
tests/                    Test suite
output/                   Default output directory

Output Format

All CLI commands output JSON to stdout. Progress messages go to stderr. Use --output to save artifacts to files.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free