Skip to main content
ClaudeWave
Skill963 estrellas del repoactualizado 4d ago

notebooklm-connector

The NotebookLM Connector automates creation, source management, and content generation within NotebookLM using browser automation through the Claude Chrome extension. Use this skill when you need to programmatically create notebooks, add multiple URL or document sources, or trigger generation of mindmaps, audio overviews, and briefing documents without manual NotebookLM interface interaction. It requires the Claude Chrome extension and active NotebookLM authentication.

Instalar en Claude Code
Copiar
git clone --depth 1 https://github.com/mohitagw15856/pm-claude-skills /tmp/notebooklm-connector && cp -r /tmp/notebooklm-connector/plugins/pm-cross/skills/notebooklm-connector ~/.claude/skills/notebooklm-connector
Después abre una sesión nueva de Claude Code; el skill carga automáticamente.

SKILL.md

# NotebookLM Connector

## The Problem

NotebookLM is one of the best AI research tools — but it doesn't connect to your other tools. Every notebook requires manual setup inside the NotebookLM UI: open browser, name the notebook, paste URLs one by one, click generate. For researchers, builders, or anyone who works with a high volume of sources, this friction compounds fast.

This skill automates NotebookLM from Claude Code using browser automation via the Claude Chrome extension.

## Prerequisites

| Requirement | Details |
|-------------|---------|
| Claude Chrome extension | Must be installed and active in your Chrome browser |
| NotebookLM account | Active account at notebooklm.google.com |
| Chrome browser | Open and signed into NotebookLM |

If the Chrome extension is not installed, this skill cannot function. There is no fallback — you will need to perform actions manually.

## Required Inputs

| Input | Required | Notes |
|-------|----------|-------|
| Action(s) to perform | Yes | What you want done — see Supported Actions below |
| Notebook name | Conditional | Required for create; optional for add/generate if a notebook is already open |
| Sources | Conditional | Required for add sources action — URLs, file paths, or pasted text |
| Output type | Conditional | Required for generate action — mindmap, audio overview, or briefing doc |

## Supported Actions

| Action | What It Does |
|--------|-------------|
| Create notebook | Opens NotebookLM, creates a new notebook with the specified title |
| Add sources | Adds one or more URLs, files, or text blocks as sources to a notebook |
| Generate mindmap | Triggers mindmap generation from the notebook's sources |
| Generate audio overview | Requests an audio overview (note: takes several minutes to render) |
| Generate briefing doc | Requests a briefing document or slide deck from sources |
| List notebooks | Lists your existing notebooks and their source counts |
| Open notebook | Navigates to a specific existing notebook by name |

Actions can be chained in a single request: "Create a notebook called 'AI Trends Q2', add these 3 URLs as sources, then generate a mindmap."

## Output Structure

After completing actions, Claude returns a structured confirmation:

```
## NotebookLM — Actions Completed

**Notebook:** [Notebook name]
**URL:** [Direct link to the notebook]
**Actions completed:**
- [x] Created notebook: "[Name]"
- [x] Added source: [URL or file name]
- [x] Added source: [URL or file name]
- [x] Triggered: Mindmap generation

**Status:** [Any pending items — e.g. "Audio overview is generating, check back in 5–10 minutes"]

**Notes:** [Any issues encountered or deviations from the requested actions]
```

If an action fails, the failed step is marked with `[ ]` and a reason is provided. See Error Handling below.

## Instructions for Claude

### Step 1 — Parse and confirm the request

Before opening any browser, parse the full request into discrete steps:

1. What notebook is being targeted (new or existing)?
2. What sources need to be added (list each URL or file)?
3. What outputs need to be generated?

If anything is ambiguous — e.g. "add my research sources" without specifying what they are — ask for clarification before proceeding. Do not guess at source URLs.

### Step 2 — Check the Chrome extension is available

Confirm browser automation is available via the Claude Chrome extension. If it is not active, stop and report:

> "This skill requires the Claude Chrome extension to be installed and active. Please install it at [extension URL] and try again."

### Step 3 — Navigate to NotebookLM

Open or navigate to `https://notebooklm.google.com`. Confirm the user is logged in. If a login screen appears, stop and ask the user to log in manually, then retry.

### Step 4 — Execute actions in order

Execute each action in the sequence requested. After each action, confirm it completed before moving to the next. Do not batch actions speculatively.

**Creating a notebook:**
- Click "New Notebook"
- Enter the specified title
- Confirm the notebook is created and visible

**Adding a URL source:**
- In the notebook, click "Add Source"
- Select "Website" or "URL"
- Paste the URL
- Wait for the source to process and appear in the sources list
- Confirm before adding the next source

**Adding pasted text:**
- Click "Add Source"
- Select "Copied text" or "Paste text"
- Paste the content
- Confirm the source appears

**Generating a mindmap:**
- Navigate to the notebook's output options
- Select "Mindmap" from available outputs
- Trigger generation
- Confirm the mindmap begins rendering

**Generating an audio overview:**
- Navigate to output options
- Select "Audio Overview"
- Trigger generation
- Note: rendering takes several minutes — report this to the user, do not wait for completion

### Step 5 — Compile and return the confirmation

Return the structured output described in the Output Structure section above, including the direct notebook URL and a checklist of completed/failed actions.

## Error Handling

If any step fails, do the following:

1. Stop at the failed step (do not attempt to continue)
2. Report the exact step that failed and what was observed
3. Suggest a manual workaround for that step
4. Offer to retry from that point

**Common failures and workarounds:**

| Failure | Likely Cause | Manual Workaround |
|---------|-------------|-------------------|
| Extension not detected | Extension not installed or disabled | Install from Chrome Web Store |
| Login screen appears | Session expired | Log in manually, then retry |
| Source fails to process | URL is paywalled or blocked | Download content and add as pasted text instead |
| Mindmap not available | Source volume too low | Add more sources (NotebookLM requires minimum content) |
| Audio overview grayed out | Sources not yet indexed | Wait 1–2 minutes for indexing, then retry |

## Limitations

- **Chrome extension required** — This skill does not work in the Claude web interfac
ai-ethics-reviewSkill

Conduct a structured ethical review of an AI or ML feature, model, or product. Use when preparing to deploy an AI system, assessing algorithmic risk, auditing a model for bias, or producing a responsible AI impact assessment. Produces a structured ethics review covering fairness, transparency, privacy, safety, accountability, and societal impact with a risk tier score, pre-deployment checklist, and prioritised mitigations.

ai-product-canvasSkill

Structure AI and ML product decisions with the rigour of any product decision. Use when building AI-powered features, evaluating LLM integrations, designing AI products, or assessing AI readiness. Produces a complete AI product canvas covering problem definition, model approach, data requirements, evaluation framework, UX design, responsible AI checklist, and launch monitoring plan.

design-handoff-briefSkill

Transform feature briefs into structured design briefs that give designers the context they need before opening Figma. Use when asked to write a design brief, create a design handoff, brief a designer on a new feature, or translate a PRD into design requirements. Produces a brief with user goal, emotional context, success criteria, constraints, edge cases, and out-of-scope boundaries.

experiment-designerSkill

Design statistically rigorous A/B tests and interpret experiment results. Use when asked to design an experiment, run an A/B test, calculate sample size, interpret test results, or assess whether an experiment was successful. Produces a complete experiment design with hypothesis, sample size, run time, success criteria, and risk flags — or a results interpretation with ship/iterate/kill recommendation.

multi-source-signal-synthesiserSkill

Synthesises user signals from multiple research sources into a unified, weighted insight brief. Use when you have data from interviews, support tickets, NPS verbatims, app reviews, or sales calls and need to reconcile contradictions, surface the underlying need behind requests, or answer 'what are users really telling us'. Produces ranked insights with confidence ratings, source weighting rationale, divergent signal analysis by user segment, and a research gap identification section.

data-analysis-standardSkill

Structure a product data analysis, metric deep-dive, funnel analysis, or cohort study. Use when asked to analyse product metrics, investigate a drop in conversion, explain a data change to stakeholders, or find the root cause of a metric movement. Produces a structured analysis with question, root cause, confidence level, and recommended action.

product-health-analysisSkill

Interpret product metrics against goals and surface actionable signals. Use when asked to analyse product health, review key metrics, investigate a performance issue, produce a health report, or assess product-market fit signals. Produces a structured health report with RAG status, trend analysis, root cause hypotheses, and prioritised actions.

retention-analysisSkill

Structure a retention analysis, churn investigation, or engagement deep-dive for any product team. Use when asked to analyse user retention, investigate churn, measure DAU/MAU, or build a retention improvement plan. Produces a retention snapshot with root cause hypotheses, aha-moment correlation, and prioritised interventions.