A high-efficiency distillation tool that converts raw URLs, PDFs, or text into structured knowledge cards, mind maps, and Q&A formats.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install kb-digest
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 kb-digest 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 kb-digest?
kb-digest is a sophisticated knowledge extraction skill designed to help users synthesize vast amounts of information into actionable formats. By utilizing advanced LLM integration through Openclaw Skills, it processes diverse inputs ranging from academic papers to technical blogs, transforming them into structured summaries that are easy to digest and store in personal knowledge bases. It effectively bridges the gap between deep reading and rapid information retrieval.
Whether you are a developer looking to summarize a new framework's documentation or a researcher managing multiple arXiv papers, this skill automates the heavy lifting of content organization. It supports multiple output modalities, ensuring that the final data fits your specific workflow, whether that involves visual mind mapping or text-based flashcards.
kb-digest Use Cases
- Summarizing complex research papers and technical documentation for rapid review.
- Converting long-form web articles into structured study cards for personal knowledge management.
- Generating mind maps to visualize the hierarchy of information in a new subject area.
- Creating Q&A pairs to assist in learning and information retention.
- Producing podcast or dialogue scripts based on technical content for alternative content consumption.
How kb-digest Works
- The skill receives an input source, which can be a URL, a local PDF file, or raw text input.
- It utilizes specialized Python libraries like pypdf and markdownify to extract and clean the content from the source.
- The extracted content is sent to a Large Language Model using the credentials and configurations provided by Openclaw Skills.
- The LLM processes the content based on a selected template (e.g., summary, mindmap, or podcast script).
- The distilled information is then formatted as a structured knowledge card and can be saved locally, displayed, or pushed to a webhook.
kb-digest Setup
To get started with kb-digest, ensure you have Python 3 installed and follow these steps:
# Navigate to the skill directory
cd /path/to/kb-digest
# Install the required dependencies
pip install -r requirements.txt
# Run the distiller on a target URL
python handler.py --url "https://example.com/article"
Environment variables for the LLM are automatically injected when running through Openclaw Skills, so no manual API key configuration is required for basic operation.
kb-digest Data Schema & Taxonomy
The skill organizes distilled data into a structured format designed for clarity. The following table describes the typical structure of a generated knowledge card:
| Component | Description |
|---|---|
| Knowledge Card Title | The primary identifier of the source material. |
| One-Liner | A single-sentence summary of the core value proposition. |
| Key Points | A numbered or bulleted list of essential takeaways. |
| Q&A Section | Frequently asked questions derived from the content. |
| Mind Map | A hierarchical Markdown representation of the content structure. |
| Metadata | Source URL, timestamp, and related reading links. |
kb-digest Advanced Features
- Multi-Format Outputs: Choose between standard cards, mind maps, Q&A pairs, summaries, or podcast scripts.
- Batch Processing: Handle multiple URLs simultaneously by providing a text file with one link per line.
- External Integration: Support for pushing processed results directly to Feishu (Lark) via webhooks.
- PDF Parsing: Dedicated logic for extracting text from research papers and multi-page documents.
- Seamless LLM Integration: Built to work natively with Openclaw Skills environment variable injection for simplified devops workflows.
SKILL.md
Loading
Related Openclaw Skills
Core Capabilities Skill
An integrated suite providing Obsidian and Git synchronization, persistent SQLite memory, and natural language query tools.
Relic
Relic is a specialized tool for capturing, distilling, and evolving a durable model of a user's identity, values, and goals for AI agents.
SafetyNet
A specialized security framework for AI agents that enforces safety protocols and ethical boundaries before high-risk system operations.
Cartier Knowledge Assistant
An intelligent encyclopedia skill providing comprehensive data on Cartier's brand story, product lines, and global market positioning.
Screen Life
An automated macOS digital life logger that generates insightful behavior reports and productivity metrics with zero configuration.
SkillBoss
SkillBoss acts as a centralized financial and functional layer for AI agents, providing a single entry point to 350+ premium tools and models.
Featured*
