nblm

内容来源：SKILL.md（标准 Skill 格式） · 原始地址 · 查看安装指南

原始内容

---
name: nblm
description: Use this skill to query your Google NotebookLM notebooks directly from Claude Code for source-grounded, citation-backed answers from Gemini. Browser automation, library management, persistent auth. Drastically reduced hallucinations through document-only responses.
---

# NotebookLM Quick Commands

Query Google NotebookLM for source-grounded, citation-backed answers.

## Environment

All dependencies and authentication are handled automatically by `run.py`:
- First run creates `.venv` and installs Python/Node.js dependencies
- If Google auth is missing or expired, a browser window opens automatically
- No manual pre-flight steps required

---

## Usage

`/nblm <command> [args]`

## Commands

### Notebook Management
| Command | Description |
|---------|-------------|
| `login` | Authenticate with Google |
| `status` | Show auth and library status |
| `accounts` | List all Google accounts |
| `accounts add` | Add a new Google account |
| `accounts switch <id>` | Switch active account (by index or email) |
| `accounts remove <id>` | Remove a Google account |
| `accounts use <id>` | Set agent-specific active account (OpenClaw isolation) |
| `accounts clear` | Clear agent-specific account override |
| `local` | List notebooks in local library |
| `remote` | List all notebooks from NotebookLM API |
| `create <name>` | Create a new notebook |
| `delete [--id ID]` | Delete a notebook |
| `rename <name> [--id ID]` | Rename a notebook |
| `summary [--id ID]` | Get AI-generated summary |
| `describe [--id ID]` | Get description and suggested topics |
| `add <url-or-id>` | Add notebook to local library (auto-detects URL vs notebook ID) |
| `activate <id>` | Set active notebook |

### Source Management
| Command | Description |
|---------|-------------|
| `sources [--id ID]` | List sources in notebook |
| `upload <file>` | Upload a single file |
| `upload <folder>` | Sync a folder of files to NotebookLM |
| `upload-zlib <url>` | Download from Z-Library and upload |
| `upload-url <url>` | Add URL as source |
| `upload-youtube <url>` | Add YouTube video as source |
| `upload-text <title> [--content TEXT]` | Add text as source |
| `source-text <source-id>` | Get full indexed text |
| `source-guide <source-id>` | Get AI summary and keywords |
| `source-rename <source-id> <name>` | Rename a source |
| `source-refresh <source-id>` | Re-fetch URL content |
| `source-delete <source-id>` | Delete a source |

**Upload options:**
- `--use-active` - Upload to the currently active notebook
- `--create-new` - Create a new notebook named after the file/folder
- `--notebook-id <id>` - Upload to a specific notebook
- `--dry-run` - Show sync plan without executing (folder sync)
- `--rebuild` - Force rebuild tracking file (folder sync)

**Important:** When user runs upload without specifying a target, ASK them first:
> "Would you like to upload to the active notebook, or create a new notebook?"
Then pass the appropriate flag (`--use-active` or `--create-new`).

### Chat & Audio/Media
| Command | Description |
|---------|-------------|
| `ask <question>` | Query NotebookLM |
| `podcast [--instructions TEXT]` | Generate audio podcast |
| `podcast-status <task-id>` | Check podcast generation status |
| `podcast-download [output-path]` | Download latest podcast |
| `briefing [--instructions TEXT]` | Generate brief audio summary |
| `debate [--instructions TEXT]` | Generate debate-style audio |
| `slides [--instructions TEXT]` | Generate slide deck |
| `slides-download [output-path]` | Download slide deck as PDF |
| `infographic [--instructions TEXT]` | Generate infographic |
| `infographic-download [output-path]` | Download infographic |
| `media-list [--type TYPE]` | List generated media (audio/video/slides/infographic) |
| `media-delete <id>` | Delete a generated media item |

## Command Routing

Based on `$ARGUMENTS`, execute the appropriate command:

$IF($ARGUMENTS,
  Parse the command from: "$ARGUMENTS"

  **login** → `python scripts/run.py auth_manager.py setup --service google`

  **accounts** → `python scripts/run.py auth_manager.py accounts list`

  **accounts add** → `python scripts/run.py auth_manager.py accounts add`

  **accounts switch <id>** → `python scripts/run.py auth_manager.py accounts switch "<id>"`

  **accounts remove <id>** → `python scripts/run.py auth_manager.py accounts remove "<id>"`

  **accounts use <id>** → `python scripts/run.py auth_manager.py accounts use "<id>"`

  **accounts clear** → `python scripts/run.py auth_manager.py accounts clear`

  **status** → Run both:
  - `python scripts/run.py auth_manager.py status`
  - `python scripts/run.py notebook_manager.py list`

  **local** → `python scripts/run.py notebook_manager.py list`

  **remote** → `python scripts/run.py nblm_cli.py notebooks`

  **create <name>** → `python scripts/run.py nblm_cli.py create "<name>"`

  **delete [--id ID]** → `python scripts/run.py nblm_cli.py delete <args>`

  **rename <name> [--id ID]** → `python scripts/run.py nblm_cli.py rename "<name>" <args>`

  **summary [--id ID]** → `python scripts/run.py nblm_cli.py summary <args>`

  **describe [--id ID]** → `python scripts/run.py nblm_cli.py describe <args>`

  **add <url-or-id>** → Smart add workflow (auto-detects URL vs notebook ID)

  **activate <id>** → `python scripts/run.py notebook_manager.py activate --id "<id>"`

  **sources [--id ID]** → `python scripts/run.py nblm_cli.py sources <args>`

  **upload <file>** → First ASK user: "Upload to active notebook or create new?" Then:
    - Active: `python scripts/run.py source_manager.py add --file "<file>" --use-active`
    - New: `python scripts/run.py source_manager.py add --file "<file>" --create-new`

  **upload <folder>** → Sync a folder:
    - First ASK user: "Sync to active notebook, create new, or specify notebook?"
    - Active: `python scripts/run.py source_manager.py sync "<folder>" --use-active`
    - New: `python scripts/run.py source_manager.py sync "<folder>" --create-new`
    - Specific: `python scripts/run.py source_manager.py sync "<folder>" --notebook-id ID`
    - Dry-run: `python scripts/run.py source_manager.py sync "<folder>" --dry-run`
    - Rebuild: `python scripts/run.py source_manager.py sync "<folder>" --rebuild`

  **upload-zlib <url>** → First ASK user: "Upload to active notebook or create new?" Then:
    - Active: `python scripts/run.py source_manager.py add --url "<url>" --use-active`
    - New: `python scripts/run.py source_manager.py add --url "<url>" --create-new`

  **upload-url <url>** → `python scripts/run.py nblm_cli.py upload-url "<url>"`

  **upload-youtube <url>** → `python scripts/run.py nblm_cli.py upload-youtube "<url>"`

  **upload-text <title>** → `python scripts/run.py nblm_cli.py upload-text "<title>" <args>`

  **source-text <id>** → `python scripts/run.py nblm_cli.py source-text "<id>"`

  **source-guide <id>** → `python scripts/run.py nblm_cli.py source-guide "<id>"`

  **source-rename <id> <name>** → `python scripts/run.py nblm_cli.py source-rename "<id>" "<name>"`

  **source-refresh <id>** → `python scripts/run.py nblm_cli.py source-refresh "<id>"`

  **source-delete <id>** → `python scripts/run.py nblm_cli.py source-delete "<id>"`

  **ask <question>** → `python scripts/run.py nblm_cli.py ask "<question>"`

  **podcast** → `python scripts/run.py artifact_manager.py generate --format DEEP_DIVE <args>`

  **podcast-status <task-id>** → `python scripts/run.py artifact_manager.py status --task-id "<task-id>"`

  **podcast-download [output-path]** → `python scripts/run.py artifact_manager.py download "<output-path>"`

  **briefing** → `python scripts/run.py artifact_manager.py generate --format BRIEF <args>`

  **debate** → `python scripts/run.py artifact_manager.py generate --format DEBATE <args>`

  **slides** → `python scripts/run.py artifact_manager.py generate-slides <args>`

  **slides-download [output-path]** → `python scripts/run.py artifact_manager.py download "<output-path>" --type slide-deck`

  **infographic** → `python scripts/run.py artifact_manager.py generate-infographic <args>`

  **infographic-download [output-path]** → `python scripts/run.py artifact_manager.py download "<output-path>" --type infographic`

  **media-list [--type TYPE]** → `python scripts/run.py artifact_manager.py list <args>`

  **media-delete <id>** → `python scripts/run.py artifact_manager.py delete "<id>"`

  If command not recognized, show usage help.,

  Show available commands with `/nblm` (no arguments)
)

## Podcast Options

```
/nblm podcast --length DEFAULT --wait --output ./podcast.mp3
/nblm podcast --instructions "Focus on the key findings"
/nblm briefing --wait --output ./summary.mp3
/nblm debate --instructions "Compare the two approaches"
```

| Option | Values |
|--------|--------|
| `--length` | `SHORT`, `DEFAULT`, `LONG` |
| `--instructions` | Custom instructions for the content |
| `--wait` | Wait for generation to complete |
| `--output` | Download path (requires `--wait`) |

## Slide Deck Options

```
/nblm slides --format DETAILED_DECK --wait --output ./presentation.pdf
/nblm slides --instructions "Focus on key diagrams" --format PRESENTER_SLIDES
```

| Option | Values |
|--------|--------|
| `--format` | `DETAILED_DECK`, `PRESENTER_SLIDES` |
| `--length` | `SHORT`, `DEFAULT` |
| `--instructions` | Custom instructions for the content |
| `--wait` | Wait for generation to complete |
| `--output` | Download path (requires `--wait`) |

## Infographic Options

```
/nblm infographic --orientation LANDSCAPE --wait --output ./visual.png
/nblm infographic --instructions "Highlight comparison" --detail-level DETAILED
```

| Option | Values |
|--------|--------|
| `--orientation` | `LANDSCAPE`, `PORTRAIT`, `SQUARE` |
| `--detail-level` | `CONCISE`, `STANDARD`, `DETAILED` |
| `--instructions` | Custom instructions for the content |
| `--wait` | Wait for generation to complete |
| `--output` | Download path (requires `--wait`) |

## Media Generation

| Command | Description | Output |
|---------|-------------|--------|
| `/nblm podcast` | Deep-dive audio discussion | MP3 |
| `/nblm briefing` | Brief audio summary | MP3 |
| `/nblm debate` | Debate-style audio | MP3 |
| `/nblm slides` | Slide deck presentation | PDF |
| `/nblm infographic` | Visual infographic | PNG |

### Examples
```
/nblm podcast --wait --output ./deep-dive.mp3
/nblm briefing --instructions "Focus on chapter 3" --wait
/nblm debate --length LONG --wait --output ./debate.mp3
/nblm slides --instructions "Include key diagrams" --format DETAILED_DECK --wait --output ./presentation.pdf
/nblm infographic --orientation LANDSCAPE --detail-level DETAILED --wait --output ./summary.png
```

### Download & Manage
```
/nblm podcast-download ./my-podcast.mp3
/nblm slides-download ./presentation.pdf
/nblm infographic-download ./visual.png
/nblm media-list                     # List all generated media
/nblm media-list --type audio        # List only audio
/nblm media-delete <id>              # Delete a media item
```

---

# Extended Documentation

## When to Use This Skill

Trigger when user:
- Mentions NotebookLM explicitly
- Shares NotebookLM URL (`https://notebooklm.google.com/notebook/...`)
- Asks to query their notebooks/documentation
- Wants to add documentation to NotebookLM library
- Uses phrases like "ask my NotebookLM", "check my docs", "query my notebook"

## ⚠️ CRITICAL: Add Command - Smart Discovery

The add command now **automatically discovers metadata** from the notebook:

```bash
# Smart Add (auto-discovers name, description, topics)
python scripts/run.py notebook_manager.py add <notebook-id-or-url>

# With optional overrides
python scripts/run.py notebook_manager.py add <id> --name "Custom Name" --topics "custom,topics"
```

**What Smart Add does:**
1. Fetches notebook title from NotebookLM API
2. Queries the notebook content to generate description and topics
3. Adds to local library with discovered metadata

**Supported input formats:**
- Notebook ID: `5fd9f36b-8000-401d-a7a0-7aa3f7832644`
- Full URL: `https://notebooklm.google.com/notebook/5fd9f36b-8000-401d-a7a0-7aa3f7832644`

NEVER manually specify `--name`, `--description`, or `--topics` unless the user explicitly provides them.

## Critical: Always Use run.py Wrapper

**NEVER call scripts directly. ALWAYS use `python scripts/run.py [script]`:**

```bash
# ✅ CORRECT - Always use run.py:
python scripts/run.py auth_manager.py status
python scripts/run.py notebook_manager.py list
python scripts/run.py ask_question.py --question "..."

# ❌ WRONG - Never call directly:
python scripts/auth_manager.py status  # Fails without venv!
```

The `run.py` wrapper automatically:
1. Creates `.venv` if needed
2. Installs all dependencies
3. Activates environment
4. Executes script properly

## Core Workflow

### Step 1: Check Authentication Status
```bash
python scripts/run.py auth_manager.py status
```

If not authenticated, proceed to setup.

### Step 2: Authenticate (One-Time Setup)
```bash
# Browser MUST be visible for manual Google login
python scripts/run.py auth_manager.py setup
```

**Important:**
- Browser is VISIBLE for authentication
- Browser window opens automatically
- User must manually log in to Google
- Tell user: "A browser window will open for Google login"

### Step 3: Manage Notebook Library

```bash
# List all notebooks
python scripts/run.py notebook_manager.py list

# BEFORE ADDING: Ask user for metadata if unknown!
# "What does this notebook contain?"
# "What topics should I tag it with?"

# Add notebook to library (ALL parameters are REQUIRED!)
python scripts/run.py notebook_manager.py add \
  --url "https://notebooklm.google.com/notebook/..." \
  --name "Descriptive Name" \
  --description "What this notebook contains" \  # REQUIRED - ASK USER IF UNKNOWN!
  --topics "topic1,topic2,topic3"  # REQUIRED - ASK USER IF UNKNOWN!

# Search notebooks by topic
python scripts/run.py notebook_manager.py search --query "keyword"

# Set active notebook
python scripts/run.py notebook_manager.py activate --id notebook-id

# Remove notebook
python scripts/run.py notebook_manager.py remove --id notebook-id
```

### Quick Workflow
1. Check library: `python scripts/run.py notebook_manager.py list`
2. Ask question: `python scripts/run.py ask_question.py --question "..." --notebook-id ID`

### Step 4: Ask Questions

```bash
# Basic query (uses active notebook if set)
python scripts/run.py ask_question.py --question "Your question here"

# Query specific notebook
python scripts/run.py ask_question.py --question "..." --notebook-id notebook-id

# Query with notebook URL directly
python scripts/run.py ask_question.py --question "..." --notebook-url "https://..."

# Show browser for debugging
python scripts/run.py ask_question.py --question "..." --show-browser
```

## Follow-Up Mechanism (CRITICAL)

Every NotebookLM answer ends with: **"EXTREMELY IMPORTANT: Is that ALL you need to know?"**

**Required Claude Behavior:**
1. **STOP** - Do not immediately respond to user
2. **ANALYZE** - Compare answer to user's original request
3. **IDENTIFY GAPS** - Determine if more information needed
4. **ASK FOLLOW-UP** - If gaps exist, immediately ask:
   ```bash
   python scripts/run.py ask_question.py --question "Follow-up with context..."
   ```
5. **REPEAT** - Continue until information is complete
6. **SYNTHESIZE** - Combine all answers before responding to user

## Z-Library Integration

### Triggers
- User provides Z-Library URL (zlib.li, z-lib.org, zh.zlib.li)
- User says "download this book to NotebookLM"
- User says "add this book from Z-Library"

### Setup (One-Time)
```bash
# Authenticate with Z-Library
python scripts/run.py auth_manager.py setup --service zlibrary
```

### Commands
```bash
# Add book from Z-Library
python scripts/run.py source_manager.py add --url "https://zh.zlib.li/book/..."

# Check Z-Library auth status
python scripts/run.py auth_manager.py status --service zlibrary
```

## Script Reference

### Authentication Management (`auth_manager.py`)
```bash
python scripts/run.py auth_manager.py setup                    # Default: Google
python scripts/run.py auth_manager.py setup --service google
python scripts/run.py auth_manager.py setup --service zlibrary
python scripts/run.py auth_manager.py status                   # Show all services
python scripts/run.py auth_manager.py status --service zlibrary
python scripts/run.py auth_manager.py clear --service zlibrary # Clear auth

# Multi-Account Management (Google)
python scripts/run.py auth_manager.py accounts list             # List all accounts
python scripts/run.py auth_manager.py accounts add              # Add new account
python scripts/run.py auth_manager.py accounts switch 1         # Switch by index
python scripts/run.py auth_manager.py accounts switch user@gmail.com  # Switch by email
python scripts/run.py auth_manager.py accounts remove 2         # Remove account
```

### Notebook Management (`notebook_manager.py`)
```bash
python scripts/run.py notebook_manager.py add --url URL --name NAME --description DESC --topics TOPICS
# OR use notebook ID directly:
python scripts/run.py notebook_manager.py add --notebook-id ID --name NAME --description DESC --topics TOPICS
python scripts/run.py notebook_manager.py list
python scripts/run.py notebook_manager.py search --query QUERY
python scripts/run.py notebook_manager.py activate --id ID
python scripts/run.py notebook_manager.py remove --id ID
python scripts/run.py notebook_manager.py stats
```

### Question Interface (`ask_question.py`)
```bash
python scripts/run.py ask_question.py --question "..." [--notebook-id ID] [--notebook-url URL] [--show-browser]
```

### Source Manager (`source_manager.py`)
```bash
# Upload to active notebook
python scripts/run.py source_manager.py add --file "/path/to/book.pdf" --use-active

# Create new notebook for upload
python scripts/run.py source_manager.py add --file "/path/to/book.pdf" --create-new

# Upload to specific notebook
python scripts/run.py source_manager.py add --file "/path/to/book.pdf" --notebook-id NOTEBOOK_ID

# Z-Library download and upload
python scripts/run.py source_manager.py add --url "https://zh.zlib.li/book/..." --use-active
python scripts/run.py source_manager.py add --url "https://zh.zlib.li/book/..." --create-new

# Sync a folder (new!)
python scripts/run.py source_manager.py sync "/path/to/docs" --use-active
python scripts/run.py source_manager.py sync "/path/to/docs" --create-new
python scripts/run.py source_manager.py sync "/path/to/docs" --notebook-id NOTEBOOK_ID

# Sync options (new!)
python scripts/run.py source_manager.py sync "/path/to/docs" --dry-run    # Preview only
python scripts/run.py source_manager.py sync "/path/to/docs" --rebuild   # Force re-hash all files
```

**Folder Sync:**
- Scans folder for supported types: PDF, TXT, MD, DOCX, HTML, EPUB
- Tracks sync state internally (no per-folder tracking file to manage)
- Sync strategy: add new, update modified (delete + re-upload), skip unchanged
- Multi-account aware (tracks which Google account was used)
**Note:** One of `--use-active`, `--create-new`, or `--notebook-id` is REQUIRED.
Uploads wait for NotebookLM processing and print progress as `Ready: N/T`. Press Ctrl+C to stop waiting.
Local file uploads use browser automation and require Google authentication.
If browser automation is unavailable, set `NOTEBOOKLM_UPLOAD_MODE=text` to upload extracted text instead (PDFs require `pypdf`).

### Data Cleanup (`cleanup_manager.py`)
```bash
python scripts/run.py cleanup_manager.py                    # Preview cleanup
python scripts/run.py cleanup_manager.py --confirm          # Execute cleanup
python scripts/run.py cleanup_manager.py --preserve-library # Keep notebooks
```

### Watchdog Status (`auth_manager.py`)
```bash
python scripts/run.py auth_manager.py watchdog-status
```

## Environment Management

The virtual environment is automatically managed:
- First run creates `.venv` automatically
- Dependencies install automatically
- Node.js dependencies install automatically
- agent-browser daemon starts on demand and keeps browser state in memory
- daemon stops after 10 minutes of inactivity (any agent-browser command resets the timer)
- set `AGENT_BROWSER_OWNER_PID` to auto-stop when the agent process exits
- `scripts/run.py` sets `AGENT_BROWSER_OWNER_PID` to its parent PID by default
- Everything isolated in skill directory

Manual setup (only if automatic fails):
```bash
python -m venv .venv
source .venv/bin/activate  # Linux/Mac
pip install -r requirements.txt
npm install
npm run install-browsers
```

## Data Storage

All data stored in `~/.claude/skills/notebooklm/data/`:
- `library.json` - Notebook metadata (with account associations)
- `auth/google/` - Multi-account Google auth
  - `index.json` - Account index (active account, list)
  - `<n>-<email>.json` - Per-account credentials
- `auth/zlibrary.json` - Z-Library auth state
- `agent_browser/session_id` - Current daemon session ID
- `agent_browser/last_activity.json` - Last activity timestamp for idle shutdown
- `agent_browser/watchdog.pid` - Idle watchdog process ID

**Security:** Protected by `.gitignore`, never commit to git.

## Configuration

Optional `.env` file in skill directory:
```env
HEADLESS=false           # Browser visibility
SHOW_BROWSER=false       # Default browser display
STEALTH_ENABLED=true     # Human-like behavior
TYPING_WPM_MIN=160       # Typing speed
TYPING_WPM_MAX=240
DEFAULT_NOTEBOOK_ID=     # Default notebook
```

## Decision Flow

```
User mentions NotebookLM
    ↓
Check auth → python scripts/run.py auth_manager.py status
    ↓
If not authenticated → python scripts/run.py auth_manager.py setup
    ↓
Check/Add notebook → python scripts/run.py notebook_manager.py list/add (with --description)
    ↓
Activate notebook → python scripts/run.py notebook_manager.py activate --id ID
    ↓
Ask question → python scripts/run.py ask_question.py --question "..."
    ↓
See "Is that ALL you need?" → Ask follow-ups until complete
    ↓
Synthesize and respond to user
```

## Troubleshooting

| Problem | Solution |
|---------|----------|
| ModuleNotFoundError | Use `run.py` wrapper |
| Authentication fails | Browser must be visible for setup! --show-browser |
| DAEMON_UNAVAILABLE | Ensure Node.js/npm installed, run `npm install`, retry |
| AUTH_REQUIRED | Run `python scripts/run.py auth_manager.py setup` |
| ELEMENT_NOT_FOUND | Verify notebook URL and re-run with fresh page load |
| Rate limit (50/day) | Wait or add another Google account with `accounts add` |
| Browser crashes | `python scripts/run.py cleanup_manager.py --preserve-library` |
| Notebook not found | Check with `notebook_manager.py list` |

## Best Practices

1. **Always use run.py** - Handles environment automatically
2. **Check auth first** - Before any operations
3. **Follow-up questions** - Don't stop at first answer
4. **Browser visible for auth** - Required for manual login
5. **Include context** - Each question is independent
6. **Synthesize answers** - Combine multiple responses

## Limitations

- No session persistence (each question = new browser)
- Rate limits on free Google accounts (50 queries/day per account; use multiple accounts to increase)
- Manual upload required (user must add docs to NotebookLM)
- Browser overhead (few seconds per question)

## Resources (Skill Structure)

**Important directories and files:**

- `scripts/` - All automation scripts (ask_question.py, notebook_manager.py, etc.)
- `data/` - Local storage for authentication and notebook library
- `references/` - Extended documentation:
  - `api_reference.md` - Detailed API documentation for all scripts
  - `troubleshooting.md` - Common issues and solutions
  - `usage_patterns.md` - Best practices and workflow examples
- `.venv/` - Isolated Python environment (auto-created on first run)
- `.gitignore` - Protects sensitive data from being committed

name: nblm description: Use this skill to query your Google NotebookLM notebooks directly from Claude Code for source-grounded, citation-backed answers from Gemini. Browser automation, library management, persistent auth. Drastically reduced hallucinations through document-only responses.

NotebookLM Quick Commands

Query Google NotebookLM for source-grounded, citation-backed answers.

Environment

All dependencies and authentication are handled automatically by run.py:

First run creates .venv and installs Python/Node.js dependencies
If Google auth is missing or expired, a browser window opens automatically
No manual pre-flight steps required

Usage

/nblm <command> [args]

Commands

Notebook Management

Command	Description
`login`	Authenticate with Google
`status`	Show auth and library status
`accounts`	List all Google accounts
`accounts add`	Add a new Google account
`accounts switch <id>`	Switch active account (by index or email)
`accounts remove <id>`	Remove a Google account
`accounts use <id>`	Set agent-specific active account (OpenClaw isolation)
`accounts clear`	Clear agent-specific account override
`local`	List notebooks in local library
`remote`	List all notebooks from NotebookLM API
`create <name>`	Create a new notebook
`delete [--id ID]`	Delete a notebook
`rename <name> [--id ID]`	Rename a notebook
`summary [--id ID]`	Get AI-generated summary
`describe [--id ID]`	Get description and suggested topics
`add <url-or-id>`	Add notebook to local library (auto-detects URL vs notebook ID)
`activate <id>`	Set active notebook

Source Management

Command	Description
`sources [--id ID]`	List sources in notebook
`upload <file>`	Upload a single file
`upload <folder>`	Sync a folder of files to NotebookLM
`upload-zlib <url>`	Download from Z-Library and upload
`upload-url <url>`	Add URL as source
`upload-youtube <url>`	Add YouTube video as source
`upload-text <title> [--content TEXT]`	Add text as source
`source-text <source-id>`	Get full indexed text
`source-guide <source-id>`	Get AI summary and keywords
`source-rename <source-id> <name>`	Rename a source
`source-refresh <source-id>`	Re-fetch URL content
`source-delete <source-id>`	Delete a source

Upload options:

--use-active - Upload to the currently active notebook
--create-new - Create a new notebook named after the file/folder
--notebook-id <id> - Upload to a specific notebook
--dry-run - Show sync plan without executing (folder sync)
--rebuild - Force rebuild tracking file (folder sync)

Important: When user runs upload without specifying a target, ASK them first:

"Would you like to upload to the active notebook, or create a new notebook?" Then pass the appropriate flag (--use-active or --create-new).

Chat & Audio/Media

Command	Description
`ask <question>`	Query NotebookLM
`podcast [--instructions TEXT]`	Generate audio podcast
`podcast-status <task-id>`	Check podcast generation status
`podcast-download [output-path]`	Download latest podcast
`briefing [--instructions TEXT]`	Generate brief audio summary
`debate [--instructions TEXT]`	Generate debate-style audio
`slides [--instructions TEXT]`	Generate slide deck
`slides-download [output-path]`	Download slide deck as PDF
`infographic [--instructions TEXT]`	Generate infographic
`infographic-download [output-path]`	Download infographic
`media-list [--type TYPE]`	List generated media (audio/video/slides/infographic)
`media-delete <id>`	Delete a generated media item

Command Routing

Based on $ARGUMENTS, execute the appropriate command:

$IF($ARGUMENTS, Parse the command from: "$ARGUMENTS"

login → python scripts/run.py auth_manager.py setup --service google

accounts → python scripts/run.py auth_manager.py accounts list

accounts add → python scripts/run.py auth_manager.py accounts add

accounts switch → python scripts/run.py auth_manager.py accounts switch "<id>"

accounts remove → python scripts/run.py auth_manager.py accounts remove "<id>"

accounts use → python scripts/run.py auth_manager.py accounts use "<id>"

accounts clear → python scripts/run.py auth_manager.py accounts clear

status → Run both:

python scripts/run.py auth_manager.py status
python scripts/run.py notebook_manager.py list

local → python scripts/run.py notebook_manager.py list

remote → python scripts/run.py nblm_cli.py notebooks

create → python scripts/run.py nblm_cli.py create "<name>"

delete [--id ID] → python scripts/run.py nblm_cli.py delete <args>

rename [--id ID] → python scripts/run.py nblm_cli.py rename "<name>" <args>

summary [--id ID] → python scripts/run.py nblm_cli.py summary <args>

describe [--id ID] → python scripts/run.py nblm_cli.py describe <args>

add → Smart add workflow (auto-detects URL vs notebook ID)

activate → python scripts/run.py notebook_manager.py activate --id "<id>"

sources [--id ID] → python scripts/run.py nblm_cli.py sources <args>

upload → First ASK user: "Upload to active notebook or create new?" Then: - Active: python scripts/run.py source_manager.py add --file "<file>" --use-active - New: python scripts/run.py source_manager.py add --file "<file>" --create-new

upload → Sync a folder: - First ASK user: "Sync to active notebook, create new, or specify notebook?" - Active: python scripts/run.py source_manager.py sync "<folder>" --use-active - New: python scripts/run.py source_manager.py sync "<folder>" --create-new - Specific: python scripts/run.py source_manager.py sync "<folder>" --notebook-id ID - Dry-run: python scripts/run.py source_manager.py sync "<folder>" --dry-run - Rebuild: python scripts/run.py source_manager.py sync "<folder>" --rebuild

upload-zlib → First ASK user: "Upload to active notebook or create new?" Then: - Active: python scripts/run.py source_manager.py add --url "<url>" --use-active - New: python scripts/run.py source_manager.py add --url "<url>" --create-new

upload-url → python scripts/run.py nblm_cli.py upload-url "<url>"

upload-youtube → python scripts/run.py nblm_cli.py upload-youtube "<url>"

upload-text </strong> → <code>python scripts/run.py nblm_cli.py upload-text "<title>" <args></code></p> <p> <strong>source-text <id></strong> → <code>python scripts/run.py nblm_cli.py source-text "<id>"</code></p> <p> <strong>source-guide <id></strong> → <code>python scripts/run.py nblm_cli.py source-guide "<id>"</code></p> <p> <strong>source-rename <id> <name></strong> → <code>python scripts/run.py nblm_cli.py source-rename "<id>" "<name>"</code></p> <p> <strong>source-refresh <id></strong> → <code>python scripts/run.py nblm_cli.py source-refresh "<id>"</code></p> <p> <strong>source-delete <id></strong> → <code>python scripts/run.py nblm_cli.py source-delete "<id>"</code></p> <p> <strong>ask <question></strong> → <code>python scripts/run.py nblm_cli.py ask "<question>"</code></p> <p> <strong>podcast</strong> → <code>python scripts/run.py artifact_manager.py generate --format DEEP_DIVE <args></code></p> <p> <strong>podcast-status <task-id></strong> → <code>python scripts/run.py artifact_manager.py status --task-id "<task-id>"</code></p> <p> <strong>podcast-download [output-path]</strong> → <code>python scripts/run.py artifact_manager.py download "<output-path>"</code></p> <p> <strong>briefing</strong> → <code>python scripts/run.py artifact_manager.py generate --format BRIEF <args></code></p> <p> <strong>debate</strong> → <code>python scripts/run.py artifact_manager.py generate --format DEBATE <args></code></p> <p> <strong>slides</strong> → <code>python scripts/run.py artifact_manager.py generate-slides <args></code></p> <p> <strong>slides-download [output-path]</strong> → <code>python scripts/run.py artifact_manager.py download "<output-path>" --type slide-deck</code></p> <p> <strong>infographic</strong> → <code>python scripts/run.py artifact_manager.py generate-infographic <args></code></p> <p> <strong>infographic-download [output-path]</strong> → <code>python scripts/run.py artifact_manager.py download "<output-path>" --type infographic</code></p> <p> <strong>media-list [--type TYPE]</strong> → <code>python scripts/run.py artifact_manager.py list <args></code></p> <p> <strong>media-delete <id></strong> → <code>python scripts/run.py artifact_manager.py delete "<id>"</code></p> <p> If command not recognized, show usage help.,</p> <p> Show available commands with <code>/nblm</code> (no arguments) )</p> <h2>Podcast Options</h2> <pre><code>/nblm podcast --length DEFAULT --wait --output ./podcast.mp3 /nblm podcast --instructions "Focus on the key findings" /nblm briefing --wait --output ./summary.mp3 /nblm debate --instructions "Compare the two approaches" </code></pre> <table> <thead> <tr> <th>Option</th> <th>Values</th> </tr> </thead> <tbody><tr> <td><code>--length</code></td> <td><code>SHORT</code>, <code>DEFAULT</code>, <code>LONG</code></td> </tr> <tr> <td><code>--instructions</code></td> <td>Custom instructions for the content</td> </tr> <tr> <td><code>--wait</code></td> <td>Wait for generation to complete</td> </tr> <tr> <td><code>--output</code></td> <td>Download path (requires <code>--wait</code>)</td> </tr> </tbody></table> <h2>Slide Deck Options</h2> <pre><code>/nblm slides --format DETAILED_DECK --wait --output ./presentation.pdf /nblm slides --instructions "Focus on key diagrams" --format PRESENTER_SLIDES </code></pre> <table> <thead> <tr> <th>Option</th> <th>Values</th> </tr> </thead> <tbody><tr> <td><code>--format</code></td> <td><code>DETAILED_DECK</code>, <code>PRESENTER_SLIDES</code></td> </tr> <tr> <td><code>--length</code></td> <td><code>SHORT</code>, <code>DEFAULT</code></td> </tr> <tr> <td><code>--instructions</code></td> <td>Custom instructions for the content</td> </tr> <tr> <td><code>--wait</code></td> <td>Wait for generation to complete</td> </tr> <tr> <td><code>--output</code></td> <td>Download path (requires <code>--wait</code>)</td> </tr> </tbody></table> <h2>Infographic Options</h2> <pre><code>/nblm infographic --orientation LANDSCAPE --wait --output ./visual.png /nblm infographic --instructions "Highlight comparison" --detail-level DETAILED </code></pre> <table> <thead> <tr> <th>Option</th> <th>Values</th> </tr> </thead> <tbody><tr> <td><code>--orientation</code></td> <td><code>LANDSCAPE</code>, <code>PORTRAIT</code>, <code>SQUARE</code></td> </tr> <tr> <td><code>--detail-level</code></td> <td><code>CONCISE</code>, <code>STANDARD</code>, <code>DETAILED</code></td> </tr> <tr> <td><code>--instructions</code></td> <td>Custom instructions for the content</td> </tr> <tr> <td><code>--wait</code></td> <td>Wait for generation to complete</td> </tr> <tr> <td><code>--output</code></td> <td>Download path (requires <code>--wait</code>)</td> </tr> </tbody></table> <h2>Media Generation</h2> <table> <thead> <tr> <th>Command</th> <th>Description</th> <th>Output</th> </tr> </thead> <tbody><tr> <td><code>/nblm podcast</code></td> <td>Deep-dive audio discussion</td> <td>MP3</td> </tr> <tr> <td><code>/nblm briefing</code></td> <td>Brief audio summary</td> <td>MP3</td> </tr> <tr> <td><code>/nblm debate</code></td> <td>Debate-style audio</td> <td>MP3</td> </tr> <tr> <td><code>/nblm slides</code></td> <td>Slide deck presentation</td> <td>PDF</td> </tr> <tr> <td><code>/nblm infographic</code></td> <td>Visual infographic</td> <td>PNG</td> </tr> </tbody></table> <h3>Examples</h3> <pre><code>/nblm podcast --wait --output ./deep-dive.mp3 /nblm briefing --instructions "Focus on chapter 3" --wait /nblm debate --length LONG --wait --output ./debate.mp3 /nblm slides --instructions "Include key diagrams" --format DETAILED_DECK --wait --output ./presentation.pdf /nblm infographic --orientation LANDSCAPE --detail-level DETAILED --wait --output ./summary.png </code></pre> <h3>Download & Manage</h3> <pre><code>/nblm podcast-download ./my-podcast.mp3 /nblm slides-download ./presentation.pdf /nblm infographic-download ./visual.png /nblm media-list # List all generated media /nblm media-list --type audio # List only audio /nblm media-delete <id> # Delete a media item </code></pre> <hr> <h1>Extended Documentation</h1> <h2>When to Use This Skill</h2> <p>Trigger when user:</p> <ul> <li>Mentions NotebookLM explicitly</li> <li>Shares NotebookLM URL (<code>https://notebooklm.google.com/notebook/...</code>)</li> <li>Asks to query their notebooks/documentation</li> <li>Wants to add documentation to NotebookLM library</li> <li>Uses phrases like "ask my NotebookLM", "check my docs", "query my notebook"</li> </ul> <h2>⚠️ CRITICAL: Add Command - Smart Discovery</h2> <p>The add command now <strong>automatically discovers metadata</strong> from the notebook:</p> <pre><code class="language-bash"># Smart Add (auto-discovers name, description, topics) python scripts/run.py notebook_manager.py add <notebook-id-or-url> # With optional overrides python scripts/run.py notebook_manager.py add <id> --name "Custom Name" --topics "custom,topics" </code></pre> <p><strong>What Smart Add does:</strong></p> <ol> <li>Fetches notebook title from NotebookLM API</li> <li>Queries the notebook content to generate description and topics</li> <li>Adds to local library with discovered metadata</li> </ol> <p><strong>Supported input formats:</strong></p> <ul> <li>Notebook ID: <code>5fd9f36b-8000-401d-a7a0-7aa3f7832644</code></li> <li>Full URL: <code>https://notebooklm.google.com/notebook/5fd9f36b-8000-401d-a7a0-7aa3f7832644</code></li> </ul> <p>NEVER manually specify <code>--name</code>, <code>--description</code>, or <code>--topics</code> unless the user explicitly provides them.</p> <h2>Critical: Always Use run.py Wrapper</h2> <p><strong>NEVER call scripts directly. ALWAYS use <code>python scripts/run.py [script]</code>:</strong></p> <pre><code class="language-bash"># ✅ CORRECT - Always use run.py: python scripts/run.py auth_manager.py status python scripts/run.py notebook_manager.py list python scripts/run.py ask_question.py --question "..." # ❌ WRONG - Never call directly: python scripts/auth_manager.py status # Fails without venv! </code></pre> <p>The <code>run.py</code> wrapper automatically:</p> <ol> <li>Creates <code>.venv</code> if needed</li> <li>Installs all dependencies</li> <li>Activates environment</li> <li>Executes script properly</li> </ol> <h2>Core Workflow</h2> <h3>Step 1: Check Authentication Status</h3> <pre><code class="language-bash">python scripts/run.py auth_manager.py status </code></pre> <p>If not authenticated, proceed to setup.</p> <h3>Step 2: Authenticate (One-Time Setup)</h3> <pre><code class="language-bash"># Browser MUST be visible for manual Google login python scripts/run.py auth_manager.py setup </code></pre> <p><strong>Important:</strong></p> <ul> <li>Browser is VISIBLE for authentication</li> <li>Browser window opens automatically</li> <li>User must manually log in to Google</li> <li>Tell user: "A browser window will open for Google login"</li> </ul> <h3>Step 3: Manage Notebook Library</h3> <pre><code class="language-bash"># List all notebooks python scripts/run.py notebook_manager.py list # BEFORE ADDING: Ask user for metadata if unknown! # "What does this notebook contain?" # "What topics should I tag it with?" # Add notebook to library (ALL parameters are REQUIRED!) python scripts/run.py notebook_manager.py add \ --url "https://notebooklm.google.com/notebook/..." \ --name "Descriptive Name" \ --description "What this notebook contains" \ # REQUIRED - ASK USER IF UNKNOWN! --topics "topic1,topic2,topic3" # REQUIRED - ASK USER IF UNKNOWN! # Search notebooks by topic python scripts/run.py notebook_manager.py search --query "keyword" # Set active notebook python scripts/run.py notebook_manager.py activate --id notebook-id # Remove notebook python scripts/run.py notebook_manager.py remove --id notebook-id </code></pre> <h3>Quick Workflow</h3> <ol> <li>Check library: <code>python scripts/run.py notebook_manager.py list</code></li> <li>Ask question: <code>python scripts/run.py ask_question.py --question "..." --notebook-id ID</code></li> </ol> <h3>Step 4: Ask Questions</h3> <pre><code class="language-bash"># Basic query (uses active notebook if set) python scripts/run.py ask_question.py --question "Your question here" # Query specific notebook python scripts/run.py ask_question.py --question "..." --notebook-id notebook-id # Query with notebook URL directly python scripts/run.py ask_question.py --question "..." --notebook-url "https://..." # Show browser for debugging python scripts/run.py ask_question.py --question "..." --show-browser </code></pre> <h2>Follow-Up Mechanism (CRITICAL)</h2> <p>Every NotebookLM answer ends with: <strong>"EXTREMELY IMPORTANT: Is that ALL you need to know?"</strong></p> <p><strong>Required Claude Behavior:</strong></p> <ol> <li><strong>STOP</strong> - Do not immediately respond to user</li> <li><strong>ANALYZE</strong> - Compare answer to user's original request</li> <li><strong>IDENTIFY GAPS</strong> - Determine if more information needed</li> <li><strong>ASK FOLLOW-UP</strong> - If gaps exist, immediately ask:<pre><code class="language-bash">python scripts/run.py ask_question.py --question "Follow-up with context..." </code></pre> </li> <li><strong>REPEAT</strong> - Continue until information is complete</li> <li><strong>SYNTHESIZE</strong> - Combine all answers before responding to user</li> </ol> <h2>Z-Library Integration</h2> <h3>Triggers</h3> <ul> <li>User provides Z-Library URL (zlib.li, z-lib.org, zh.zlib.li)</li> <li>User says "download this book to NotebookLM"</li> <li>User says "add this book from Z-Library"</li> </ul> <h3>Setup (One-Time)</h3> <pre><code class="language-bash"># Authenticate with Z-Library python scripts/run.py auth_manager.py setup --service zlibrary </code></pre> <h3>Commands</h3> <pre><code class="language-bash"># Add book from Z-Library python scripts/run.py source_manager.py add --url "https://zh.zlib.li/book/..." # Check Z-Library auth status python scripts/run.py auth_manager.py status --service zlibrary </code></pre> <h2>Script Reference</h2> <h3>Authentication Management (<code>auth_manager.py</code>)</h3> <pre><code class="language-bash">python scripts/run.py auth_manager.py setup # Default: Google python scripts/run.py auth_manager.py setup --service google python scripts/run.py auth_manager.py setup --service zlibrary python scripts/run.py auth_manager.py status # Show all services python scripts/run.py auth_manager.py status --service zlibrary python scripts/run.py auth_manager.py clear --service zlibrary # Clear auth # Multi-Account Management (Google) python scripts/run.py auth_manager.py accounts list # List all accounts python scripts/run.py auth_manager.py accounts add # Add new account python scripts/run.py auth_manager.py accounts switch 1 # Switch by index python scripts/run.py auth_manager.py accounts switch user@gmail.com # Switch by email python scripts/run.py auth_manager.py accounts remove 2 # Remove account </code></pre> <h3>Notebook Management (<code>notebook_manager.py</code>)</h3> <pre><code class="language-bash">python scripts/run.py notebook_manager.py add --url URL --name NAME --description DESC --topics TOPICS # OR use notebook ID directly: python scripts/run.py notebook_manager.py add --notebook-id ID --name NAME --description DESC --topics TOPICS python scripts/run.py notebook_manager.py list python scripts/run.py notebook_manager.py search --query QUERY python scripts/run.py notebook_manager.py activate --id ID python scripts/run.py notebook_manager.py remove --id ID python scripts/run.py notebook_manager.py stats </code></pre> <h3>Question Interface (<code>ask_question.py</code>)</h3> <pre><code class="language-bash">python scripts/run.py ask_question.py --question "..." [--notebook-id ID] [--notebook-url URL] [--show-browser] </code></pre> <h3>Source Manager (<code>source_manager.py</code>)</h3> <pre><code class="language-bash"># Upload to active notebook python scripts/run.py source_manager.py add --file "/path/to/book.pdf" --use-active # Create new notebook for upload python scripts/run.py source_manager.py add --file "/path/to/book.pdf" --create-new # Upload to specific notebook python scripts/run.py source_manager.py add --file "/path/to/book.pdf" --notebook-id NOTEBOOK_ID # Z-Library download and upload python scripts/run.py source_manager.py add --url "https://zh.zlib.li/book/..." --use-active python scripts/run.py source_manager.py add --url "https://zh.zlib.li/book/..." --create-new # Sync a folder (new!) python scripts/run.py source_manager.py sync "/path/to/docs" --use-active python scripts/run.py source_manager.py sync "/path/to/docs" --create-new python scripts/run.py source_manager.py sync "/path/to/docs" --notebook-id NOTEBOOK_ID # Sync options (new!) python scripts/run.py source_manager.py sync "/path/to/docs" --dry-run # Preview only python scripts/run.py source_manager.py sync "/path/to/docs" --rebuild # Force re-hash all files </code></pre> <p><strong>Folder Sync:</strong></p> <ul> <li>Scans folder for supported types: PDF, TXT, MD, DOCX, HTML, EPUB</li> <li>Tracks sync state internally (no per-folder tracking file to manage)</li> <li>Sync strategy: add new, update modified (delete + re-upload), skip unchanged</li> <li>Multi-account aware (tracks which Google account was used) <strong>Note:</strong> One of <code>--use-active</code>, <code>--create-new</code>, or <code>--notebook-id</code> is REQUIRED. Uploads wait for NotebookLM processing and print progress as <code>Ready: N/T</code>. Press Ctrl+C to stop waiting. Local file uploads use browser automation and require Google authentication. If browser automation is unavailable, set <code>NOTEBOOKLM_UPLOAD_MODE=text</code> to upload extracted text instead (PDFs require <code>pypdf</code>).</li> </ul> <h3>Data Cleanup (<code>cleanup_manager.py</code>)</h3> <pre><code class="language-bash">python scripts/run.py cleanup_manager.py # Preview cleanup python scripts/run.py cleanup_manager.py --confirm # Execute cleanup python scripts/run.py cleanup_manager.py --preserve-library # Keep notebooks </code></pre> <h3>Watchdog Status (<code>auth_manager.py</code>)</h3> <pre><code class="language-bash">python scripts/run.py auth_manager.py watchdog-status </code></pre> <h2>Environment Management</h2> <p>The virtual environment is automatically managed:</p> <ul> <li>First run creates <code>.venv</code> automatically</li> <li>Dependencies install automatically</li> <li>Node.js dependencies install automatically</li> <li>agent-browser daemon starts on demand and keeps browser state in memory</li> <li>daemon stops after 10 minutes of inactivity (any agent-browser command resets the timer)</li> <li>set <code>AGENT_BROWSER_OWNER_PID</code> to auto-stop when the agent process exits</li> <li><code>scripts/run.py</code> sets <code>AGENT_BROWSER_OWNER_PID</code> to its parent PID by default</li> <li>Everything isolated in skill directory</li> </ul> <p>Manual setup (only if automatic fails):</p> <pre><code class="language-bash">python -m venv .venv source .venv/bin/activate # Linux/Mac pip install -r requirements.txt npm install npm run install-browsers </code></pre> <h2>Data Storage</h2> <p>All data stored in <code>~/.claude/skills/notebooklm/data/</code>:</p> <ul> <li><code>library.json</code> - Notebook metadata (with account associations)</li> <li><code>auth/google/</code> - Multi-account Google auth<ul> <li><code>index.json</code> - Account index (active account, list)</li> <li><code><n>-<email>.json</code> - Per-account credentials</li> </ul> </li> <li><code>auth/zlibrary.json</code> - Z-Library auth state</li> <li><code>agent_browser/session_id</code> - Current daemon session ID</li> <li><code>agent_browser/last_activity.json</code> - Last activity timestamp for idle shutdown</li> <li><code>agent_browser/watchdog.pid</code> - Idle watchdog process ID</li> </ul> <p><strong>Security:</strong> Protected by <code>.gitignore</code>, never commit to git.</p> <h2>Configuration</h2> <p>Optional <code>.env</code> file in skill directory:</p> <pre><code class="language-env">HEADLESS=false # Browser visibility SHOW_BROWSER=false # Default browser display STEALTH_ENABLED=true # Human-like behavior TYPING_WPM_MIN=160 # Typing speed TYPING_WPM_MAX=240 DEFAULT_NOTEBOOK_ID= # Default notebook </code></pre> <h2>Decision Flow</h2> <pre><code>User mentions NotebookLM ↓ Check auth → python scripts/run.py auth_manager.py status ↓ If not authenticated → python scripts/run.py auth_manager.py setup ↓ Check/Add notebook → python scripts/run.py notebook_manager.py list/add (with --description) ↓ Activate notebook → python scripts/run.py notebook_manager.py activate --id ID ↓ Ask question → python scripts/run.py ask_question.py --question "..." ↓ See "Is that ALL you need?" → Ask follow-ups until complete ↓ Synthesize and respond to user </code></pre> <h2>Troubleshooting</h2> <table> <thead> <tr> <th>Problem</th> <th>Solution</th> </tr> </thead> <tbody><tr> <td>ModuleNotFoundError</td> <td>Use <code>run.py</code> wrapper</td> </tr> <tr> <td>Authentication fails</td> <td>Browser must be visible for setup! --show-browser</td> </tr> <tr> <td>DAEMON_UNAVAILABLE</td> <td>Ensure Node.js/npm installed, run <code>npm install</code>, retry</td> </tr> <tr> <td>AUTH_REQUIRED</td> <td>Run <code>python scripts/run.py auth_manager.py setup</code></td> </tr> <tr> <td>ELEMENT_NOT_FOUND</td> <td>Verify notebook URL and re-run with fresh page load</td> </tr> <tr> <td>Rate limit (50/day)</td> <td>Wait or add another Google account with <code>accounts add</code></td> </tr> <tr> <td>Browser crashes</td> <td><code>python scripts/run.py cleanup_manager.py --preserve-library</code></td> </tr> <tr> <td>Notebook not found</td> <td>Check with <code>notebook_manager.py list</code></td> </tr> </tbody></table> <h2>Best Practices</h2> <ol> <li><strong>Always use run.py</strong> - Handles environment automatically</li> <li><strong>Check auth first</strong> - Before any operations</li> <li><strong>Follow-up questions</strong> - Don't stop at first answer</li> <li><strong>Browser visible for auth</strong> - Required for manual login</li> <li><strong>Include context</strong> - Each question is independent</li> <li><strong>Synthesize answers</strong> - Combine multiple responses</li> </ol> <h2>Limitations</h2> <ul> <li>No session persistence (each question = new browser)</li> <li>Rate limits on free Google accounts (50 queries/day per account; use multiple accounts to increase)</li> <li>Manual upload required (user must add docs to NotebookLM)</li> <li>Browser overhead (few seconds per question)</li> </ul> <h2>Resources (Skill Structure)</h2> <p><strong>Important directories and files:</strong></p> <ul> <li><code>scripts/</code> - All automation scripts (ask_question.py, notebook_manager.py, etc.)</li> <li><code>data/</code> - Local storage for authentication and notebook library</li> <li><code>references/</code> - Extended documentation:<ul> <li><code>api_reference.md</code> - Detailed API documentation for all scripts</li> <li><code>troubleshooting.md</code> - Common issues and solutions</li> <li><code>usage_patterns.md</code> - Best practices and workflow examples</li> </ul> </li> <li><code>.venv/</code> - Isolated Python environment (auto-created on first run)</li> <li><code>.gitignore</code> - Protects sensitive data from being committed</li> </ul> </div> </section> <p class="raw-footer muted">本页内容由上游仓库自动镜像，如有疑问请访问<a href="https://cdn.jsdelivr.net/gh/magicseek/nblm@main/SKILL.md" rel="noopener">原始仓库</a>。</p> </div> </div></main> <footer class="site"><div class="container"> <nav class="foot-nav"> <a href="/skills/">全部 Skill</a><a href="/roles/">角色</a><a href="/scenes/">场景</a><a href="/collections/">合集</a><a href="/favorites/">收藏</a><a href="/submit/">投稿</a><a href="/about/">关于</a><a href="/feed.xml">RSS</a><a href="/sitemap-index.xml">Sitemap</a> </nav> <p>321skill.com · AI Agent Skill 导航 · 发现好用的 Skill</p> </div></footer> <script> document.getElementById('copyAllBtn')?.addEventListener('click',async function(e){var btn=e.currentTarget;var ta=document.getElementById('rawBody');var text=ta?ta.value:'';try{await navigator.clipboard.writeText(text);}catch(err){ta&&ta.select();document.execCommand('copy');}var old=btn.textContent;btn.textContent='已复制 ✓';btn.classList.add('copied');setTimeout(function(){btn.textContent=old;btn.classList.remove('copied');},1500);}); </script> </body> </html>