DeepWiki

v0.0.10Knowledge & Memorystable

Crawls and converts DeepWiki documentation pages into Markdown format, enabling retrieval of complete documentation or specific pages from repositories. Note: Currently not functional due to DeepWiki scraping restrictions.

mcpmcp-startermodel-context-protocol
Share:
1,354
Stars
0
Downloads
0
Weekly
0/5

What is DeepWiki?

DeepWiki is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to crawls and converts deepwiki documentation pages into markdown format, enabling retrieval of complete documentation or specific pages from repositories. note: currently not functional due to deepwiki ...

Crawls and converts DeepWiki documentation pages into Markdown format, enabling retrieval of complete documentation or specific pages from repositories. Note: Currently not functional due to DeepWiki scraping restrictions.

This server falls under the Knowledge & Memory category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Crawls and converts DeepWiki documentation pages into Markdo

Use Cases

Crawl and retrieve DeepWiki documentation in Markdown format.
Access complete documentation repositories for context.
regenrek

Maintainer

LicenseMIT License
Languagetypescript
Versionv0.0.10
UpdatedMay 22, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

NPM

npx -y mcp-deepwiki

Manual Installation

npx -y mcp-deepwiki

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use DeepWiki

The DeepWiki MCP Server lets AI assistants fetch and read documentation from DeepWiki — an AI-generated wiki site that provides deep technical documentation for open-source GitHub repositories. By exposing a single 'deepwiki_fetch' tool, the server crawls DeepWiki pages and converts them into clean Markdown, giving your AI assistant rich, structured context about libraries and frameworks without manual copy-paste. It supports both single-page retrieval and aggregate multi-page crawls with configurable depth, making it useful for quickly loading an entire library's architecture overview or a specific sub-section into the AI's context window.

Prerequisites

  • Node.js 18+ installed (required to run via npx)
  • An MCP-compatible client such as Claude Desktop or Cursor
  • Network access to deepwiki.com (the tool only processes URLs from that domain)
  • The server currently has scraping restrictions imposed by DeepWiki — some repositories may be inaccessible
1

Add mcp-deepwiki to your MCP client configuration

Open your MCP client config file and add the mcp-deepwiki entry. The server runs via npx and requires no local installation.

{
  "mcpServers": {
    "mcp-deepwiki": {
      "command": "npx",
      "args": ["-y", "mcp-deepwiki@latest"]
    }
  }
}
2

Restart your MCP client

Fully quit and relaunch your client (e.g. Claude Desktop) so the server is registered and the deepwiki_fetch tool becomes available.

3

Fetch a repository's documentation in aggregate mode

Use the deepwiki_fetch tool with a deepwiki.com URL and mode 'aggregate' to retrieve all documentation for a repository as a single concatenated Markdown document.

4

Fetch a single documentation page

Request a specific sub-page of a repository's DeepWiki documentation by providing its exact URL.

5

(Optional) Tune performance via environment variables

Create a .env file to adjust concurrency, timeouts, and retry behavior for unreliable network conditions.

DEEPWIKI_MAX_CONCURRENCY=3
DEEPWIKI_REQUEST_TIMEOUT=60000
DEEPWIKI_MAX_RETRIES=5
DEEPWIKI_RETRY_DELAY=500

DeepWiki Examples

Client configuration

Complete MCP client config entry for mcp-deepwiki running via npx.

{
  "mcpServers": {
    "mcp-deepwiki": {
      "command": "npx",
      "args": ["-y", "mcp-deepwiki@latest"]
    }
  }
}

Prompts to try

Example prompts to retrieve DeepWiki documentation through your AI assistant.

- "Use deepwiki to fetch the full documentation for tailwindlabs/tailwindcss"
- "Fetch the architecture overview for shadcn-ui/ui from DeepWiki"
- "deepwiki fetch single page https://deepwiki.com/tailwindlabs/tailwindcss/2.2-theme-system"
- "Get the DeepWiki docs for vercel/next.js and summarize the routing system"
- "Fetch https://deepwiki.com/user/repo and explain how the authentication module works"

Troubleshooting DeepWiki

Tool returns an error or empty content for a repository

DeepWiki has scraping restrictions that may block certain repositories. This is a known limitation noted in the package. Try reducing DEEPWIKI_MAX_CONCURRENCY to 1 and increasing DEEPWIKI_REQUEST_TIMEOUT. Also verify the repository exists on deepwiki.com by visiting the URL in a browser.

npx takes a long time or fails on first run

The '-y' flag auto-accepts the package install. If it still fails, try installing globally first with 'npm install -g mcp-deepwiki' and changing the command in config to 'mcp-deepwiki' with no args.

Fetched content is incomplete or truncated

Use mode 'aggregate' with maxDepth set to 1 or 2 to limit crawl scope. Very large repositories may exceed context window limits — try fetching specific sub-pages instead of the full aggregate.

Frequently Asked Questions about DeepWiki

What is DeepWiki?

DeepWiki is a Model Context Protocol (MCP) server that crawls and converts deepwiki documentation pages into markdown format, enabling retrieval of complete documentation or specific pages from repositories. note: currently not functional due to deepwiki scraping restrictions. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install DeepWiki?

Install via npm with the command: npx -y mcp-deepwiki. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).

Which AI clients work with DeepWiki?

DeepWiki works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is DeepWiki free to use?

Yes, DeepWiki is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.

Browse More Knowledge & Memory MCP Servers

Explore all knowledge & memory servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "deepwiki-mcp-server": { "command": "npx", "args": ["-y", "mcp-deepwiki"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use DeepWiki?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides