notebooklm（NotebookLM 笔记研究） Skills下载安装命令和教程

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手，先确认命令和权限，再让它执行。

请帮我安装这个 Agent Skill：notebooklm（NotebookLM 笔记研究）
来源仓库：https://github.com/proyecto26/notebooklm-ai-plugin
仓库路径：skills/notebooklm
安装命令：
npx skills add https://github.com/proyecto26/notebooklm-ai-plugin --skill notebooklm
安装前请先检查当前环境是否支持对应 CLI，并向我确认将要执行的命令、安装目录、联网范围和文件读写权限；确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill；本站只展示命令，不托管安装包，也不自动执行。

skills.shnpx skills

npx skills add https://github.com/proyecto26/notebooklm-ai-plugin --skill notebooklm

简介

用于查找、检索和筛选相关信息，支持关键词快速定位候选结果。

适用于需要根据任务场景或来源线索在 Codex、Claude 等平台中获取信息时。
可通过 GitHub 仓库路径和 README 进一步核验具体用法和功能范围。
安装方式：通过 npx 从 proyecto26/notebooklm-ai-plugin 仓库添加技能。
注意：使用前请确认权限范围、维护状态及是否触发联网或命令执行操作。

SKILL.md

NotebookLM AI Plugin

Supports:

Chat with Notebook AI (source-grounded Q&A with citations)
Slide Deck generation (PDF/PPTX)
Audio Overview (M4A -- deep dive, brief, critique, debate formats)
Video Overview (MP4 -- classic, whiteboard, kawaii, anime, watercolor styles)
Mind Map (HTML)
Flashcards (HTML/JSON)
Quiz (HTML/JSON)
Infographic (PNG -- landscape, portrait, square)
Report (Markdown -- briefing doc, study guide, blog post)
Data Table (CSV / Google Sheets export)
Source management (add URLs, YouTube, files, pasted text; list, delete)
Fast/Deep web research with auto-import
Notes management (create, update, delete, list)
Notebook library management (add, list, search, activate)

Quick start

# Authenticate first (opens Chrome for Google login)
npx -y bun scripts/main.ts login

# Add a notebook to library
npx -y bun scripts/main.ts notebooks add https://notebooklm.google.com/notebook/abc123 --name "My Research"

# Generate artifacts
npx -y bun scripts/main.ts generate slide_deck --notebook abc123 --output slides.pdf
npx -y bun scripts/main.ts generate audio --format deep_dive --length long
npx -y bun scripts/main.ts generate video --style whiteboard --output explainer.mp4
npx -y bun scripts/main.ts generate quiz --difficulty medium --quantity more --json
npx -y bun scripts/main.ts generate mind_map --output mindmap.png
npx -y bun scripts/main.ts generate infographic --orientation portrait --output info.png
npx -y bun scripts/main.ts generate report --format study_guide --output report.md
npx -y bun scripts/main.ts generate flashcards --difficulty easy --json
npx -y bun scripts/main.ts generate data_table --output data.csv

Commands

Authentication

npx -y bun scripts/main.ts login

First run opens Chrome for Google login. Cookies are cached for subsequent runs.

# Force cookie refresh
npx -y bun scripts/main.ts login --force

Notebook Management

# List all notebooks in library
npx -y bun scripts/main.ts notebooks list

# Add a notebook by URL
npx -y bun scripts/main.ts notebooks add <url> [--name <name>] [--description <desc>] [--topics <t1,t2>]

# Remove a notebook from library
npx -y bun scripts/main.ts notebooks remove <id>

# Set active notebook (used as default for generation)
npx -y bun scripts/main.ts notebooks activate <id>

# Search notebooks
npx -y bun scripts/main.ts notebooks search <query>

Chat with Notebook AI

# Ask a question about your notebook sources
npx -y bun scripts/main.ts chat --question "What are the key findings?" --notebook abc123

# Continue a conversation
npx -y bun scripts/main.ts chat --question "Tell me more about that" --conversation-id <id>

# Get JSON output with citations
npx -y bun scripts/main.ts chat --question "Summarize the methodology" --json

Source Management

# List all sources in a notebook
npx -y bun scripts/main.ts sources list --notebook abc123

# Add sources
npx -y bun scripts/main.ts sources add-url https://example.com/article --notebook abc123
npx -y bun scripts/main.ts sources add-youtube https://youtube.com/watch?v=xxx --notebook abc123
npx -y bun scripts/main.ts sources add-text --title "My Notes" --content "Important findings..." --notebook abc123
npx -y bun scripts/main.ts sources add-file ./paper.pdf --notebook abc123

# Delete a source
npx -y bun scripts/main.ts sources delete <sourceId> --notebook abc123

Supported file types: PDF, TXT, MD, DOCX, CSV, EPUB, images (PNG, JPG, WEBP, etc.), audio/video files.

Research

# Fast web research (finds relevant sources quickly)
npx -y bun scripts/main.ts research fast --query "latest AI agent frameworks" --notebook abc123

# Deep research (comprehensive report with analysis)
npx -y bun scripts/main.ts research deep --query "state of LLM reasoning" --notebook abc123

# Auto-import found sources into the notebook
npx -y bun scripts/main.ts research fast --query "topic" --notebook abc123 --import

# Check research status
npx -y bun scripts/main.ts research status --notebook abc123

Notes

# List notes in a notebook
npx -y bun scripts/main.ts notes list --notebook abc123

# Create a note
npx -y bun scripts/main.ts notes create --title "Key Takeaways" --content "1. Finding one..." --notebook abc123

# Update a note
npx -y bun scripts/main.ts notes update <noteId> --title "Updated Title" --content "New content" --notebook abc123

# Delete a note
npx -y bun scripts/main.ts notes delete <noteId> --notebook abc123

Artifact Generation

npx -y bun scripts/main.ts generate <type> [options]

Types: slide_deck, audio, video, mind_map, flashcards, quiz, infographic, report, data_table

Options

Global Options

Option	Description
`--notebook <url\	id>`	Notebook URL or library ID (defaults to active notebook)
`--output <path>`	Output file path (auto-named if omitted)
`--instructions <text>`	Custom instructions for generation
`--json`	Output as JSON
`--login`	Refresh cookies only, then exit
`--help`, `-h`	Show help

Slide Deck Options

Option	Description
`--format <type>`	Output format: `pdf` (default), `pptx`

Audio Overview Options

Option	Description
`--format <type>`	Audio format: `deep_dive` (default), `brief`, `critique`, `debate`
`--length <length>`	Duration: `short`, `default`, `long`
`--language <lang>`	Language code (default: `en`)

Video Overview Options

Option	Description
`--style <style>`	Visual style: `auto` (default), `classic`, `whiteboard`, `kawaii`, `anime`, `watercolor`
`--format <type>`	Video format: `explainer` (default), `brief`

Quiz Options

Option	Description
`--difficulty <level>`	Difficulty: `easy`, `medium` (default), `hard`
`--quantity <amount>`	Number of questions: `fewer`, `standard`, `more`

Flashcards Options

Option	Description
`--difficulty <level>`	Difficulty: `easy`, `medium` (default), `hard`
`--quantity <amount>`	Number of cards: `fewer`, `standard`, `more`

Infographic Options

Option	Description
`--orientation <type>`	Layout: `landscape` (default), `portrait`, `square`

Report Options

Option	Description
`--format <type>`	Report format: `briefing_doc` (default), `study_guide`, `blog_post`

Data Table Options

Option	Description
`--format <type>`	Output format: `csv` (default), `sheets` (Google Sheets export)

Artifact Types

Type	Output Format	Description
`slide_deck`	PDF/PPTX	Presentation slides summarizing notebook sources
`audio`	M4A	Audio overview in conversation format (deep dive, brief, critique, debate)
`video`	MP4	Animated video overview with visual styles
`mind_map`	PNG	Visual mind map of key concepts and relationships
`flashcards`	HTML/JSON	Study flashcards generated from source material
`quiz`	HTML/JSON	Multiple-choice quiz with answer key
`infographic`	PNG	Visual summary infographic in various orientations
`report`	Markdown	Written report (briefing doc, study guide, blog post)
`data_table`	CSV/Sheets	Structured data extracted from sources

Authentication

First run opens Chrome to authenticate with Google. Cookies are cached for subsequent runs. Uses CDP browser automation for the login flow, then direct batchexecute RPC calls for all operations.

# Force cookie refresh
npx -y bun scripts/main.ts login --force

Environment variables

Variable	Description
`NOTEBOOKLM_DATA_DIR`	Data directory override
`NOTEBOOKLM_COOKIE_PATH`	Cookie file path
`NOTEBOOKLM_CHROME_PROFILE_DIR`	Chrome profile directory
`NOTEBOOKLM_OUTPUT_DIR`	Default output directory

Rate Limits

NotebookLM free tier limits:

Resource	Limit
Audio/video overviews	3 per day
Reports/flashcards/quizzes	10 per day
Daily chats	50
Total notebooks	100
Sources per notebook	50

Examples

Generate a slide deck from a specific notebook

npx -y bun scripts/main.ts generate slide_deck \
  --notebook https://notebooklm.google.com/notebook/abc123 \
  --output presentation.pdf

Create an audio deep dive with custom instructions

npx -y bun scripts/main.ts generate audio \
  --format deep_dive \
  --length long \
  --instructions "Focus on the methodology section and compare approaches"

Generate a whiteboard-style video overview

npx -y bun scripts/main.ts generate video \
  --style whiteboard \
  --length medium \
  --output explainer.mp4

Create a quiz and get JSON output for integration

npx -y bun scripts/main.ts generate quiz \
  --difficulty hard \
  --quantity more \
  --json > quiz_data.json

Generate a portrait infographic

npx -y bun scripts/main.ts generate infographic \
  --orientation portrait \
  --instructions "Highlight the three main findings" \
  --output summary.png

Export structured data as CSV

npx -y bun scripts/main.ts generate data_table \
  --format csv \
  --output extracted_data.csv

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明，方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用；具体调用方式请以来源页面和 README 为准。

平台分布

Codex

33.62%

按下载量换算192

Claude

29.48%

按下载量换算169

Cursor

18.28%

按下载量换算105

Gemini CLI

8.99%

按下载量换算51

安全审计

Gen Agent Trust Hub

通过

Socket

可疑

Snyk

可疑

权限和风险

敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置，应进入高风险复核队列，默认不自动发布。

安装前确认

本站仅展示第三方公开信息，不托管安装包，不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果，不能写成本站确认安全。当前只有一个来源，正式发布前建议补源仓库或其他目录站核验。

来源信息

GitHub 仓库

github.com

/proyecto26/notebooklm-ai-plugin

SKILL.md 来源

github.com

/proyecto26/notebooklm-ai-plugin/tree/main/skills/notebooklm

skills-sh

skills.sh

/proyecto26/notebooklm-ai-plugin/notebooklm

继续浏览同类 Skills

更多研究检索 Skills（73,675）返回 Skills 目录

上一个 Skill

notebooklm（NotebookLM 笔记研究）

notebooklm 用于查找、检索和筛选相关信息，适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态，以及是否会触发联网、命令执行或文件读写。

下一个 Skill

notebooklm（NotebookLM 笔记研究）

notebooklm 用于查找、检索和筛选相关信息，适合在 Local Agent 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态，以及是否会触发联网、命令执行或文件读写。