ClawKB enables AI agents to interact with a centralized knowledge base for automated registration, research capture, and data retrieval via HTTP.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install clawkb
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 clawkb 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 ClawKB?
ClawKB is a specialized interface designed for AI agents to communicate seamlessly with a ClawKB server over HTTP. As part of the Openclaw Skills ecosystem, it allows agents to perform high-level operations such as self-registration, bearer token management, and structured data manipulation. This skill acts as a persistent memory layer, enabling agents to store, search, and update information in a professional-grade knowledge management system.
By leveraging this skill, developers can bridge the gap between ephemeral agent conversations and long-term organizational knowledge. Whether it is uploading images to MinIO, managing technical reports, or creating internal wiki-style links, ClawKB provides the necessary endpoints to turn an AI agent into a sophisticated knowledge worker.
ClawKB Use Cases
- Registering unique agent identities with custom avatars for multi-agent environments.
- Automating the archival of nightly research reports and pricing data.
- Managing technical assets through direct image uploads to integrated storage.
- Cross-referencing complex documentation using internal wiki-style entry links.
- Enabling collaborative workflows through automated entry comments and status tracking.
How ClawKB Works
- Connect to the specified ClawKB server instance using a provided base URL.
- Register the agent through the authentication endpoint to generate a unique apiToken.
- Include the apiToken as a Bearer token in the Authorization header for all subsequent API requests.
- Perform data operations such as creating entries, uploading attachments, or updating existing records.
- Utilize the search API or the optional auto-recall plugin to retrieve contextually relevant knowledge during conversations.
ClawKB Setup
To integrate ClawKB into your Openclaw Skills setup, follow these steps:
- Define your base URL and register your agent:
BASE_URL="http://your-server:3500"
curl -sS "$BASE_URL/api/auth/register-agent" \
-H 'Content-Type: application/json' \
-d '{
"name": "My Agent",
"avatarUrl": "https://example.com/avatar.png"
}'
- Save the
apiTokenfrom the response. - Verify your connection:
TOKEN="clawkb_your_token"
curl -sS "$BASE_URL/api/me" -H "Authorization: Bearer $TOKEN"
- (Optional) Install the auto-recall plugin for automatic context injection:
openclaw plugins install @hata1234/clawkb-openclaw
ClawKB Data Schema & Taxonomy
ClawKB utilizes a structured schema to organize information within Openclaw Skills:
| Component | Key Attributes | Purpose |
|---|---|---|
| Entry | type, source, title, content, tags, status | The primary unit of stored knowledge. |
| Image | url, key, filename, size, mimeType | Metadata for media assets stored via MinIO. |
| Comment | content, author, entryId | Threaded feedback and notes on specific entries. |
| Reference | `[[entry:ID | Title]]` |
Entries also support a flexible metadata JSON object for storing arbitrary key-value pairs like confidence scores or regional identifiers.
ClawKB Advanced Features
- Auto-Recall Plugin Support: Automatically search and inject relevant knowledge into the agent's system prompt based on conversation context.
- Multi-Instance Parallelism: Simultaneously connect to and query personal, team, and public ClawKB instances.
- Wiki-Style Internal Linking: Use the
[[entry:ID|Display Title]]syntax to create a mesh of interconnected documentation. - Per-Sender Token Mapping: Assign different access levels to various users through server-side ACL and specific token management.
- Plugin-Backed UI Flows: Handle
pluginRenderdata to support specialized interface blocks within entry details.
SKILL.md
Loading
Related Openclaw Skills
MiniMax Vision & Search MCP
A powerful Model Context Protocol (MCP) skill for deep image understanding and real-time web searching using the MiniMax API.
py_mnn_kb — MNN Knowledge Base Skill
A local GraphRAG knowledge base that combines vector search, BM25, and knowledge graphs using MNN embeddings for high-performance RAG workflows.
XHS Rental OCR
A specialized data extraction tool that uses Apple Vision to convert Xiaohongshu notes and social media images into structured Excel or CSV files.
AI Video Script Generator
A comprehensive AI-powered tool that transforms keywords into production-ready video scripts featuring storyboards, visual prompts, and voiceover copy.
Smart Parenting Feynman
A sophisticated AI persona skill that combines the Feynman technique with a warm, playful Chinese parenting style for teaching and creative orchestration.
Custom Grok Search
A powerful search utility enabling AI agents to access real-time web data and X/Twitter discussions via xAI Grok APIs.
Featured*
