NotebookLM Research Workflow

What it does

When your agent uses it

• Building a source-grounded notebook from articles, videos and local files
• Asking cited questions about a set of sources
• Generating a podcast-style audio overview, slides or a study guide from sources
• Running web research and importing the results into a notebook

Example prompts

• “Create a notebook called AI safety evidence from these three URLs and summarize it.”
• “Run deep web research on recent empirical evaluations and import the results.”
• “Generate a quiz and flashcards from my notebook and download them.”

Requirements

• The notebooklm-py package
• A signed-in NotebookLM browser session

Workflow steps

8 steps, taken from the first numbered list in SKILL.md.

1. 1Verify authentication before a long workflow.
2. 2Confirm sources were ingested; treat partial or failed source entries honestly.
3. 3Ask focused questions and retain returned citation metadata.
4. 4Use exact notebook and artifact IDs in repeated automation.
5. 5Use bounded source/artifact concurrency; generation is quota-sensitive.
6. 6Do not delete notebooks or overwrite output without explicit user intent.
7. 7Do not claim a draft was published; no publishing integration exists here.
8. 8Return the JSON result or a faithful summary, including partial failures.

What it can do on your machine

Read from SKILL.md and the folder at commit dc38bd8. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Ships 1 file in scripts/, which the agent can run.

  Shell commands in SKILL.md call:

  • uvx

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  Hosts in commands or code, which the agent is likely to contact:

  • youtu.be

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

NotebookLM Research Workflow loads about 1.8k tokens when it runs, and up to ~4.8k if it reads all its reference files. Until then it costs about 136 tokens; SKILL.md has 499 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~136

When it runs · the whole SKILL.md, loaded when a task matches

~1.8k

With references · SKILL.md plus every file in references/, read only if the agent opens them

~4.8k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from claude-world/notebooklm-skill at commit dc38bd8, republished under its MIT licence (© claude-world). 499 words, ~1,785 tokens.

name

notebooklm-research

description

Automate source-grounded research with Google NotebookLM. Create notebooks from URLs, text, or local files; ask cited questions; run fast or deep web research; create articles and social drafts; and generate or download audio, video, cinematic video, slides, reports, study guides, quizzes, flashcards, mind maps, infographics, and data tables. Use when a user asks for NotebookLM, cited source analysis, research-to-content workflows, podcasts, slides, study material, artifact generation, RSS digests, or trend research.

notebooklm-auth setup
notebooklm-auth verify

uvx --from notebooklm-py notebooklm login

notebooklm-skill create \
  --title "AI safety evidence" \
  --sources https://example.com/article https://youtu.be/example \
  --files ./paper.pdf \
  --text-sources "A user-supplied observation" \
  --strict

notebooklm-skill list
notebooklm-skill list-sources --notebook "AI safety evidence"
notebooklm-skill summarize --notebook "AI safety evidence"
notebooklm-skill ask --notebook "AI safety evidence" --query "What findings conflict?"

notebooklm-skill add-source --notebook "AI safety evidence" --url https://example.com/new
notebooklm-skill add-source --notebook "AI safety evidence" --file ./appendix.docx
notebooklm-skill add-source --notebook "AI safety evidence" \
  --text "Raw notes" --text-title "Interview notes"

notebooklm-skill research \
  --notebook "AI safety evidence" \
  --query "Recent empirical evaluations" \
  --mode deep --max-sources 10

notebooklm-skill generate \
  --notebook "AI safety evidence" \
  --type slides --lang zh-TW \
  --slide-format presenter-slides \
  --output ./output/deck.pptx --output-format pptx

notebooklm-skill generate --notebook NOTEBOOK_ID --type audio --no-wait
notebooklm-skill list-artifacts --notebook NOTEBOOK_ID --type audio
notebooklm-skill download --notebook NOTEBOOK_ID --type audio \
  --artifact-id ARTIFACT_ID --output ./output/podcast.m4a

notebooklm-skill podcast --notebook NOTEBOOK_ID --output podcast.m4a
notebooklm-skill qa --notebook NOTEBOOK_ID --difficulty hard --output quiz.json

notebooklm-skill generate --help

notebooklm-pipeline research-to-article \
  --sources https://example.com/a https://example.com/b \
  --title "Evidence review" --language zh-TW --audience "engineers"

notebooklm-pipeline research-to-social \
  --sources https://example.com/a --platform threads --variants 3

notebooklm-pipeline batch-digest \
  --rss https://example.com/feed.xml --max-entries 20 --qa-count 5

notebooklm-pipeline generate-all \
  --files ./paper.pdf --types audio slides report mind-map \
  --output-dir ./output --artifact-concurrency 2

notebooklm-mcp

{
  "mcpServers": {
    "notebooklm": {
      "command": "uvx",
      "args": ["--from", "notebooklm-skill", "notebooklm-mcp"]
    }
  }
}

notebooklm-mcp --http --host 127.0.0.1 --port 8765

notebooklm-auth verify
notebooklm-auth setup                 # missing or expired session
notebooklm-skill list-artifacts --notebook NOTEBOOK_ID  # inspect a timed-out job

© claude-world, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 60 other files (scripts, references) in the repository root of claude-world/notebooklm-skill.

• SKILL.md
• .devcontainer/devcontainer.json
• .env.example
• .github/CODEOWNERS
• .github/ISSUE_TEMPLATE/bug_report.yml
• .github/ISSUE_TEMPLATE/config.yml
• .github/ISSUE_TEMPLATE/feature_request.yml
• .github/ISSUE_TEMPLATE/question.yml
• .github/copilot-instructions.md
• .github/dependabot.yml
• .github/workflows/ci.yml
• .github/workflows/codeql.yml
• .github/workflows/release.yml
• .gitignore
• .mcp.json
• AGENTS.md
• CHANGELOG.md
• … and 44 more

Open the folder on GitHubat commit dc38bd8