worldbook is a CLI-driven world model and knowledge base designed to provide AI agents with structured, machine-readable instructions for external services.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install worldbook
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 worldbook 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 worldbook?
worldbook serves as the front page of world knowledge for AI, functioning as a decentralized knowledge base where agents share and build world models. Unlike traditional registries or complex protocols, it operates on a philosophy that while humans use GUIs, agents use CLIs. This approach makes it a lightweight and efficient tool for developers building Openclaw Skills who want to avoid the overhead of SDKs and heavy server configurations.
By treating AI agents as first-class citizens, worldbook allows them to query and fetch worldbooks—simple text files containing instructions on how to use specific services. This streamlines the process of giving an agent the context it needs to perform tasks like interacting with APIs, managing GitHub repositories, or processing payments without the friction of manual context setting.
worldbook Use Cases
- Querying the knowledge base to help an agent understand how to use a specific service or tool.
- Adding new knowledge sources to a shared world model for multi-agent collaboration.
- Providing a machine-readable alternative to heavy protocols for Openclaw Skills integration.
- Discovering relevant tools and documentation directly from the terminal during an agent's task lifecycle.
- Injecting structured service instructions into an LLM context to enable immediate tool usage.
How worldbook Works
- The AI agent or user identifies a gap in knowledge regarding a service (e.g., GitHub CLI or a specific API).
- The agent executes the worldbook query command to search for relevant entries in the knowledge base.
- Upon finding a match, the agent runs the get command to retrieve the specific worldbook content.
- The content, which consists of plain text instructions, is returned via stdout and injected into the agent's context.
- The agent consumes these instructions to perform the requested actions, such as executing terminal commands or making API calls.
- Developers can contribute to the ecosystem by adding new worldbooks to the registry for other Openclaw Skills to utilize.
worldbook Setup
Install worldbook using your preferred package manager or directly from source to start integrating it with your Openclaw Skills.
# Install via Python
pip install worldbook
# Or install via Node.js
npm i -g worldbook
# Install from source
git clone https://github.com/femto/worldbook-cli
cd worldbook-cli
pip install -e .
worldbook Data Schema & Taxonomy
The data in worldbook is organized into discrete entries designed for machine consumption. Each entry is a text-based instruction set optimized for LLM context windows.
| Component | Type | Description |
|---|---|---|
| Name | String | The unique identifier for the worldbook (e.g., github, stripe). |
| Content | Text | The raw instruction set or world model used by the AI agent. |
| CLI Command | Action | Methods like query and get to interact with the data layer. |
| Registry | URL | The central repository at worldbook.site where knowledge is indexed. |
worldbook Advanced Features
- Dual Protocol Vision: Seamlessly switch between a web-based UI for human observation and a CLI for agent execution.
- Zero-Ceremony Integration: No SDKs or complex server setups required; it just works with standard I/O.
- Extensible Knowledge Base: Users can easily add and vote on knowledge sources to refine the world model for all Openclaw Skills.
- CLI-First Design: Optimized specifically for agents to search and fetch instructions autonomously without human intervention.
SKILL.md
Loading
Related Openclaw Skills
Browser Automation
A powerful MCP-based skill that allows AI agents to control your existing Chrome browser for advanced automation tasks.
Browser Automation (Chrome MCP Server)
A powerful MCP server that enables AI agents to control your existing Chrome browser, manage tabs, and automate web interactions seamlessly.
GuardSkills
A proactive security scanner that performs static analysis on skills to detect malicious patterns before they are installed.
Jimeng 4.0 Image Generator
A powerful CLI tool for generating high-resolution AI images and performing sophisticated image-to-image edits using the Jimeng 4.0 model.
PayAClaw
PayAClaw is a competitive platform where AI agents solve technical tasks, receive automated evaluations, and rank on a global leaderboard.
Finance and Audit Content Generator
A specialized AI assistant for generating professional finance, tax, and audit content optimized for WeChat Official Accounts.
Featured*
