Menu
Access NotebookLM to ask questions and retrieve information. Specify a NotebookLM URL and query.
The review-first prompt is selected by default. You can switch to a direct command or download a local copy.
Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
A direct command skips the review prompt. Inspect the source before running it.
npx skills add https://github.com/thirdlf03/ccpeek --skill notebooklm-queryThe command stays on one line. Scroll horizontally to inspect it before copying.
Prefer a local copy? Download the files currently available to SkillsMP.
Based on SOC occupation classification
Perform a comprehensive code quality review
TypeScript and Next.js coding standards. Auto-triggered when writing, reviewing, or refactoring TypeScript/React code. Covers type patterns, component patterns, Biome rules, and import conventions.
React and Next.js frontend design patterns. Auto-triggered when creating React components, implementing data fetching, managing state, or optimizing performance. Covers Server Components, Suspense, streaming, and rendering.
Showing SKILL.md
| name | notebooklm-query |
| description | Access NotebookLM to ask questions and retrieve information. Specify a NotebookLM URL and query. |
| argument-hint | <NotebookLM URL> <question or research topic> |
| allowed-tools | ["mcp__claude-in-chrome__tabs_context_mcp","mcp__claude-in-chrome__tabs_create_mcp","mcp__claude-in-chrome__navigate","mcp__claude-in-chrome__computer","mcp__claude-in-chrome__read_page","mcp__claude-in-chrome__find","mcp__claude-in-chrome__form_input","mcp__claude-in-chrome__javascript_tool","mcp__claude-in-chrome__get_page_text","Read","Glob","Grep","AskUserQuestion"] |
Access a NotebookLM notebook, ask questions via chat, and retrieve information.
The user provides:
https://notebooklm.google.com/notebook/xxxx)If arguments are missing, use AskUserQuestion to clarify.
tabs_context_mcp to get tab infotabs_create_mcp to create a new tab and navigate to the URLscreenshot to check the page stateCraft an effective question from the user's research topic and send it to NotebookLM.
find to locate the chat input fieldform_input to type the question textfind to locate the "Send" button and click the one near the chat input
screenshot to confirm the response is displayedjavascript_tool to extract the response text:
const chatArea = document.querySelector('[role="log"], [class*="chat"], [class*="conversation"]');
if (chatArea) {
const text = chatArea.innerText;
const lastQ = text.lastIndexOf('<partial question text>');
if (lastQ >= 0) text.substring(lastQ, lastQ + 5000);
}
Present the retrieved information in this format:
## NotebookLM Research Results
### Source
- Notebook: [Title](URL)
- Sources: N items
### Q1: <question>
<response summary>
### Q2: <question>
<response summary>
### Key Takeaways
- Point 1
- Point 2
- ...
javascript_tool with varying offsets for multiple retrieval passestabs_context_mcp and re-access in a new tabread_page to list interactive elements and identify the button