A professional tool for the segmented generation, modification, and management of Jupyter Notebooks with advanced features like embedded grading and media.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install notebook-builder
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install notebook-builder using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Notebook Builder?
Notebook Builder is a specialized skill designed for developers and educators who need to programmatically create or refine Jupyter Notebooks (.ipynb). It solves the common problem of LLM timeouts and content truncation by employing a segmented generation strategy, allowing for the stable creation of large, complex documents.
Beyond simple cell creation, this skill provides a robust helper library to manage the entire notebook lifecycle. It enables the seamless integration of local images via base64 encoding directly into Markdown cells and features a secure, hash-based quiz system for interactive learning. As a key component of the Openclaw Skills ecosystem, it ensures technical accuracy and state consistency by executing all operations through a controlled terminal environment.
Notebook Builder Use Cases
- Creating comprehensive programming courseware and technical tutorials with structured chapters.
- Automating the generation of data analysis reports with embedded visualizations.
- Building interactive coding exercises with secure, automated grading systems.
- Merging multiple disparate notebooks into a single, cohesive master document.
- Refactoring existing notebooks by reordering, inserting, or deleting specific cells based on keyword searches.
How Notebook Builder Works
- Initialize a new notebook object using the helper library to define specific kernel metadata.
- Plan the notebook structure by breaking content into 3-5 logical segments to prevent generation errors.
- Execute segmented cell creation where the agent generates Markdown and Code cells for one section at a time.
- Append segments to the notebook and persist the state using incremental saves to the local file system.
- Verify the current state after each operation by calling summary functions to check cell count and content integrity.
- Finalize the document by adding automated tables of contents, summary grading cells, or exporting to pure Python scripts.
Notebook Builder Setup
To use this skill, ensure the helper scripts are accessible to your agent. All operations must be executed via the terminal tool to maintain process state.
# Navigate to your skill directory
cd <skill_script_directory>
# The agent will typically initialize the environment with:
import sys
sys.path.insert(0, "<skill_script_directory>")
from nb_helpers import *
Notebook Builder Data Schema & Taxonomy
The skill manages data through a structured dictionary representing the Jupyter Notebook format. Key components include:
| Component | Description |
|---|---|
| Cells | Individual units of Markdown or Code with associated metadata and tags. |
| Metadata | Stores kernel information, cell IDs, and SHA256 hashes for quiz answers. |
| Assets | Images are converted to base64 strings for direct embedding within the .ipynb file. |
| Tags | Utilizes the standard Jupyter metadata.tags taxonomy for grouping and filtering cells. |
Notebook Builder Advanced Features
- Segmented generation to bypass LLM context limits and prevent output truncation.
- Base64 image embedding for standalone notebooks without external dependencies.
- Secure hashing system (SHA256) for quiz answers to prevent students from viewing solutions in plain text.
- Automated Table of Contents (TOC) generation based on Markdown header levels.
- Multi-notebook merging with optional divider cells and source attribution.
- Targeted cell manipulation using keyword or unique ID positioning.
SKILL.md
Loading
Related Openclaw Skills
Podwise Podcast Copilot
A powerful podcast intelligence tool that searches transcripts, summarizes episodes, and converts audio or video into structured AI insights.
Railroaded
Railroaded is an autonomous D&D platform where AI agents act as players in a world governed by real dice rolls and persistent consequences.
Feishu @ Mention (feishu-at)
A specialized skill for ensuring @ mentions in Feishu trigger real notifications through rich text post formatting.
Skill SEO Optimizer
A specialized tool for auditing and optimizing AI coding agent skills to improve their discovery and search ranking across major skill directories.
Success Plan Generator
A specialized AI skill for B2B Customer Success teams to generate structured success plans, track milestones, and identify renewal signals.
Web Interface Interaction Optimizer
A specialized toolkit for building high-performance, flicker-free web dashboards designed to monitor AI agent activities in real-time.
Featured*
