A powerful tool to interact with GitHub repository documentation and get AI-grounded answers about codebases.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install deepwiki-mcp-skill
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 deepwiki-mcp-skill 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 DeepWiki MCP Skill?
The DeepWiki MCP Skill provides a streamlined interface for developers to interact with GitHub documentation using AI-powered queries. By integrating with the Model Context Protocol (MCP), this skill enables users to ask specific questions about any indexed repository, retrieve the wiki structure, or read documentation contents directly from the CLI. It serves as an essential utility for those using Openclaw Skills to automate codebase research and technical exploration.
Built on the robust DeepWiki engine, the skill ensures that responses are technically accurate and grounded in the actual repository data. Whether you are troubleshooting an obscure API or trying to understand the architecture of a new project, this skill bridges the gap between raw code and actionable intelligence.
DeepWiki MCP Skill Use Cases
- Rapidly onboarding developers to complex open-source or private GitHub projects.
- Searching for specific API implementations or usage patterns within a codebase.
- Summarizing repository documentation to understand project goals and structures.
- Validating documentation accuracy against the actual codebase state.
How DeepWiki MCP Skill Works
- The user installs the uxc CLI tool to manage MCP connections.
- A persistent link is established between the local environment and the DeepWiki MCP endpoint.
- The user triggers a tool such as ask_question by providing a repository name and a natural language query.
- The DeepWiki engine processes the request, searches the indexed repository data, and returns a grounded response in a JSON envelope.
- The Openclaw Skills runner extracts the relevant text for the developer or AI agent.
DeepWiki MCP Skill Setup
Prerequisites
- Ensure uxc is installed and available in your PATH.
- Repository must be indexed on DeepWiki (https://deepwiki.com).
Installation
Install uxc via Homebrew:
brew tap holon-run/homebrew-tap
brew install uxc
Link the DeepWiki skill to create a local CLI command:
uxc link deepwiki-mcp-cli mcp.deepwiki.com/mcp
Verification
deepwiki-mcp-cli -h
DeepWiki MCP Skill Data Schema & Taxonomy
The skill utilizes a standard request/response pattern via MCP. The primary inputs and outputs are structured as follows:
| Parameter | Type | Description |
|---|---|---|
| repoName | String | The GitHub repository in owner/repo format (e.g., facebook/react). |
| question | String | The natural language query for the codebase documentation. |
Output Parsing:
Results are returned in an MCP JSON envelope. The human-readable text can be extracted from the path: .data.content[].text.
DeepWiki MCP Skill Advanced Features
- Multi-repository querying: Support for querying up to 10 repositories in a single session for comparative analysis.
- Direct CLI Aliasing: Use uxc link to create custom shorthand commands for specific repository interactions.
- Grounded AI Responses: Unlike general LLMs, responses are strictly grounded in the indexed documentation to prevent hallucinations.
- Fallback Execution: Supports direct uxc calls if the linked command is unavailable, ensuring high availability within Openclaw Skills workflows.
SKILL.md
Loading
Related Openclaw Skills
Context7 MCP Skill
A high-performance bridge for AI agents to query live, version-accurate documentation for virtually any software library.
CPA Update
A comprehensive maintenance tool for the CLI Proxy API designed to handle secure image upgrades, configuration changes, and robust rollback procedures.
Daily Report Writer
An automated tool to generate and store professional Markdown work reports based on your daily achievements and blockers.
Daily Report (joig)
A productivity skill for generating structured Chinese daily reports in Markdown format to streamline personal reflection and progress tracking.
Dune MCP Skill
A powerful tool for AI agents to discover, query, and visualize blockchain data directly from Dune Analytics.
Etherscan MCP Skill
A high-performance interface for onchain investigation, providing deep access to Ethereum blockchain data via the Etherscan MCP.
Featured*
