Run any Skill in Manus with one click

$pwd:

notebooklm-internal

Name: Notebooklm Internal
Author: techwavedev

// INTERNAL ONLY. Forked write-capable variant of the public notebooklm skill. Adds programmatic source ingestion (add_source) for the X/YouTube → NotebookLM pipeline (#119). Headless, non-interactive, cron-driven. NEVER ship publicly.

Run Skill in Manus

$ git log --oneline --stat

stars:3

forks:2

updated:April 15, 2026 at 09:48

File Explorer

22 files

SKILL.md

readonly

name	notebooklm-internal
description	INTERNAL ONLY. Forked write-capable variant of the public notebooklm skill. Adds programmatic source ingestion (add_source) for the X/YouTube → NotebookLM pipeline (#119). Headless, non-interactive, cron-driven. NEVER ship publicly.
risk	high
source	internal-fork
forked_from	skills/notebooklm
internal	true
public	false
non_interactive	true
date_added	2026-04-13

NotebookLM Internal (Write-Capable) Skill

⚠️ INTERNAL USE ONLY — excluded from public release via .private. This fork adds programmatic source ingestion (issues #126 ensure_notebook, #127 add_source) for the X/YouTube → NotebookLM pipeline. All browser automation runs headless by default and is designed to execute under cron with zero human interaction. The first-time auth_manager.py setup is the only step that can open a visible browser; everything else (queries, source ingestion, notebook resolution) is fully non-interactive.

Original Skill Documentation (forked from `skills/notebooklm/`)

Query Google NotebookLM for source-grounded answers from your uploaded documents. Each question opens a fresh browser session, retrieves the answer, and closes.

When to Use

Trigger when user:

Mentions NotebookLM or shares a NotebookLM URL
Asks to query their notebooks/documentation
Uses phrases like "ask my NotebookLM", "check my docs", "query my notebook"

Critical: Always Use run.py Wrapper

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

# ✅ CORRECT:
python scripts/run.py auth_manager.py status
python scripts/run.py notebook_manager.py list

# ❌ WRONG (fails without venv):
python scripts/auth_manager.py status

The wrapper auto-creates .venv, installs deps, and activates the environment.

Core Workflow

1. Check Auth

python scripts/run.py auth_manager.py status

If not authenticated → python scripts/run.py auth_manager.py setup (browser opens for Google login).

2. Manage Notebooks

python scripts/run.py notebook_manager.py list
python scripts/run.py notebook_manager.py add --url URL --name NAME --description DESC --topics TOPICS
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

3. Smart Add (when user doesn't provide details)

Query the notebook first to discover its content, then add with accurate metadata:

# Step 1: Discover content
python scripts/run.py ask_question.py --question "What is the content of this notebook? What topics are covered?" --notebook-url "[URL]"

# Step 2: Add with discovered info
python scripts/run.py notebook_manager.py add --url "[URL]" --name "[discovered]" --description "[discovered]" --topics "[discovered]"

NEVER guess descriptions — use Smart Add or ask the user.

4. Ask Questions

python scripts/run.py ask_question.py --question "Your question" [--notebook-id ID] [--notebook-url URL] [--show-browser]

Follow-Up Protocol

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

STOP — Don't immediately respond to user
ANALYZE — Compare answer to user's original request
IDENTIFY GAPS — Determine if more information needed
ASK FOLLOW-UP — If gaps exist, query again with context
SYNTHESIZE — Combine all answers before responding to user

Decision Flow

User mentions NotebookLM
  → Check auth (auth_manager.py status)
  → If not auth'd → setup (browser visible)
  → Check/Add notebook (notebook_manager.py list/add)
  → Activate notebook (notebook_manager.py activate --id ID)
  → Ask question (ask_question.py --question "...")
  → Follow-up until complete
  → Synthesize and respond

Data Storage

All data stored locally in data/ within the skill directory:

library.json — Notebook metadata
auth_info.json — Authentication status
browser_state/ — Browser cookies and session

Protected by .gitignore — never committed to git.

Limitations

No session persistence (each question = new browser session)
Rate limits on free Google accounts (~50 queries/day)
Manual upload required (user adds docs to NotebookLM web UI)
Browser overhead (few seconds per question)

References

For detailed information, see:

references/api_reference.md — Full script API, arguments, and examples
references/troubleshooting.md — Common issues and solutions
references/usage_patterns.md — Best practices and workflow patterns
references/configuration.md — Environment variables and .env setup

AGI Framework Integration

Adapted for @techwavedev/agi-agent-kit

Memory-First Protocol

python3 execution/memory_manager.py auto --query "NotebookLM patterns and prior queries"

Storing Results

python3 execution/memory_manager.py store \
  --content "NotebookLM: <what was queried/decided>" \
  --type technical --project <project> \
  --tags notebooklm documentation

Ingest Pipeline Integration

The ingest pipeline uses execution/source_fetchers to collect content and execution/ingest_dispatcher.py to orchestrate fetch → dedup → format → upload. See directives/sources_to_notebooklm.md for the full SOP.

Multi-Agent Collaboration

python3 execution/cross_agent_context.py store \
  --agent "<your-agent>" \
  --action "NotebookLM query completed — <summary>" \
  --project <project>

related-skills.json

same repository

claude-code-design.md

from "techwavedev/agi-agent-kit"

Answer questions about Claude Code / Anthropic agent design patterns (CLAUDE.md layering, skills anatomy, progressive disclosure, memory-first, orchestration, sub-agents, worktree isolation, Karpathy loop). Primary backend is a NotebookLM RAG over Anthropic design notebooks; falls back to bundled references when the notebook is not registered.

2026-04-153

webcrawler.md

from "techwavedev/agi-agent-kit"

Documentation harvesting agent for crawling and extracting content from documentation websites. Use for crawling documentation sites and extracting all pages about a subject, building offline knowledge bases from online docs, harvesting API references, tutorials, or guides from documentation portals, creating structured markdown exports from multi-page documentation, and downloading and organizing technical docs for embedding or RAG pipelines. Supports recursive crawling with depth control, content filtering, and structured output.

2026-04-123

ui-ux-pro-max.md

from "techwavedev/agi-agent-kit"

UI/UX design intelligence. 50 styles, 21 palettes, 50 font pairings, 20 charts, 9 stacks (React, Next.js, Vue, Svelte, SwiftUI, React Native, Flutter, Tailwind, shadcn/ui). Actions: plan, build, cr...

2026-04-123

image-ai-generator.md

from "techwavedev/agi-agent-kit"

Generates images via Openrouter API using AI image models. Supports two modes: test (cheap model for iteration) and production (high-quality model for final output). Handles prompt construction, API calls, base64 decoding, and file saving. Supports reference images (logos, mascots) for brand-consistent generation.

2026-04-123

qdrant-memory.md

from "techwavedev/agi-agent-kit"

Intelligent token optimization through Qdrant-powered semantic caching and long-term memory. Use for (1) Semantic Cache - avoid LLM calls entirely for semantically similar queries with 100% token savings, (2) Long-Term Memory - retrieve only relevant context chunks instead of full conversation history with 80-95% context reduction, (3) Hybrid Search - combine vector similarity with keyword filtering for technical queries, (4) Memory Management - store and retrieve conversation memories, decisions, and code patterns with metadata filtering. Triggers when needing to cache responses, remember past interactions, optimize context windows, or implement RAG patterns.

2026-04-123

activecampaign-automation.md

from "techwavedev/agi-agent-kit"

Automate ActiveCampaign tasks via Rube MCP (Composio): manage contacts, tags, list subscriptions, automation enrollment, and tasks. Always search tools first for current schemas.

2026-04-123

package.json

"author": "techwavedev"

"repository": "techwavedev/agi-agent-kit"

View GitHub Repository View Creator Repositories

$ install --global

$ download --local

Run Skill in Manus

$ useful --forSOC

Software DevelopersComputer and Mathematical Occupations15-1252L4

name	notebooklm-internal
description	INTERNAL ONLY. Forked write-capable variant of the public notebooklm skill. Adds programmatic source ingestion (add_source) for the X/YouTube → NotebookLM pipeline (#119). Headless, non-interactive, cron-driven. NEVER ship publicly.
risk	high
source	internal-fork
forked_from	skills/notebooklm
internal	true
public	false
non_interactive	true
date_added	2026-04-13

NotebookLM Internal (Write-Capable) Skill

⚠️ INTERNAL USE ONLY — excluded from public release via .private. This fork adds programmatic source ingestion (issues #126 ensure_notebook, #127 add_source) for the X/YouTube → NotebookLM pipeline. All browser automation runs headless by default and is designed to execute under cron with zero human interaction. The first-time auth_manager.py setup is the only step that can open a visible browser; everything else (queries, source ingestion, notebook resolution) is fully non-interactive.

Original Skill Documentation (forked from `skills/notebooklm/`)

Query Google NotebookLM for source-grounded answers from your uploaded documents. Each question opens a fresh browser session, retrieves the answer, and closes.

When to Use

Trigger when user:

Mentions NotebookLM or shares a NotebookLM URL
Asks to query their notebooks/documentation
Uses phrases like "ask my NotebookLM", "check my docs", "query my notebook"

Critical: Always Use run.py Wrapper

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

# ✅ CORRECT:
python scripts/run.py auth_manager.py status
python scripts/run.py notebook_manager.py list

# ❌ WRONG (fails without venv):
python scripts/auth_manager.py status

The wrapper auto-creates .venv, installs deps, and activates the environment.

Core Workflow

1. Check Auth

python scripts/run.py auth_manager.py status

If not authenticated → python scripts/run.py auth_manager.py setup (browser opens for Google login).

2. Manage Notebooks

python scripts/run.py notebook_manager.py list
python scripts/run.py notebook_manager.py add --url URL --name NAME --description DESC --topics TOPICS
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

3. Smart Add (when user doesn't provide details)

Query the notebook first to discover its content, then add with accurate metadata:

# Step 1: Discover content
python scripts/run.py ask_question.py --question "What is the content of this notebook? What topics are covered?" --notebook-url "[URL]"

# Step 2: Add with discovered info
python scripts/run.py notebook_manager.py add --url "[URL]" --name "[discovered]" --description "[discovered]" --topics "[discovered]"

NEVER guess descriptions — use Smart Add or ask the user.

4. Ask Questions

python scripts/run.py ask_question.py --question "Your question" [--notebook-id ID] [--notebook-url URL] [--show-browser]

Follow-Up Protocol

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

STOP — Don't immediately respond to user
ANALYZE — Compare answer to user's original request
IDENTIFY GAPS — Determine if more information needed
ASK FOLLOW-UP — If gaps exist, query again with context
SYNTHESIZE — Combine all answers before responding to user

Decision Flow

User mentions NotebookLM
  → Check auth (auth_manager.py status)
  → If not auth'd → setup (browser visible)
  → Check/Add notebook (notebook_manager.py list/add)
  → Activate notebook (notebook_manager.py activate --id ID)
  → Ask question (ask_question.py --question "...")
  → Follow-up until complete
  → Synthesize and respond

Data Storage

All data stored locally in data/ within the skill directory:

library.json — Notebook metadata
auth_info.json — Authentication status
browser_state/ — Browser cookies and session

Protected by .gitignore — never committed to git.

Limitations

No session persistence (each question = new browser session)
Rate limits on free Google accounts (~50 queries/day)
Manual upload required (user adds docs to NotebookLM web UI)
Browser overhead (few seconds per question)

References

For detailed information, see:

references/api_reference.md — Full script API, arguments, and examples
references/troubleshooting.md — Common issues and solutions
references/usage_patterns.md — Best practices and workflow patterns
references/configuration.md — Environment variables and .env setup

AGI Framework Integration

Adapted for @techwavedev/agi-agent-kit

Memory-First Protocol

python3 execution/memory_manager.py auto --query "NotebookLM patterns and prior queries"

Storing Results

python3 execution/memory_manager.py store \
  --content "NotebookLM: <what was queried/decided>" \
  --type technical --project <project> \
  --tags notebooklm documentation

Ingest Pipeline Integration

Multi-Agent Collaboration

python3 execution/cross_agent_context.py store \
  --agent "<your-agent>" \
  --action "NotebookLM query completed — <summary>" \
  --project <project>

notebooklm-internal

NotebookLM Internal (Write-Capable) Skill

Original Skill Documentation (forked from skills/notebooklm/)

When to Use

Critical: Always Use run.py Wrapper

Core Workflow

1. Check Auth

2. Manage Notebooks

3. Smart Add (when user doesn't provide details)

4. Ask Questions

Follow-Up Protocol

Decision Flow

Data Storage

Limitations

References

AGI Framework Integration

Memory-First Protocol

Storing Results

Ingest Pipeline Integration

Multi-Agent Collaboration

More from this repository

More from this repository

NotebookLM Internal (Write-Capable) Skill

Original Skill Documentation (forked from skills/notebooklm/)

When to Use

Critical: Always Use run.py Wrapper

Core Workflow

1. Check Auth

2. Manage Notebooks

3. Smart Add (when user doesn't provide details)

4. Ask Questions

Follow-Up Protocol

Decision Flow

Data Storage

Limitations

References

AGI Framework Integration

Memory-First Protocol

Storing Results

Ingest Pipeline Integration

Multi-Agent Collaboration

Original Skill Documentation (forked from `skills/notebooklm/`)

Original Skill Documentation (forked from `skills/notebooklm/`)