MCP Documentation Server
Provides comprehensive access to MCP documentation through structured guides, full-text search, and interactive development workflows for building servers and clients.
README
MCP Documentation Server 📚
A Model Context Protocol (MCP) server that provides comprehensive access to MCP documentation, guides, and development workflows through a standardized interface.
Features ✨
This MCP server provides comprehensive documentation access through:
🔧 Tools
- mcp_docs_guide: Structured guides for MCP development topics
- search_docs: Full-text search across all MCP documentation
- get_docs_by_category: Browse documentation by category
💬 Prompts
Interactive workflows for MCP development:
- mcp_quick_start: Get started quickly with MCP development
- server_development_workflow: Complete server development workflow
- client_integration_guide: Guide for integrating MCP into applications
- troubleshooting_workflow: Systematic troubleshooting guide
- deployment_guide: Production deployment best practices
📄 Resources
Direct access to individual documentation files:
- mcp-docs://filename.md: Access any documentation file directly
- Organized by category with rich metadata
- Full markdown content with proper formatting
Installation 📦
NPM Installation
npm install -g mcp-docs-server
Using with Claude Desktop
Add to your Claude Desktop configuration (~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"mcp-docs": {
"command": "npx",
"args": ["-y", "mcp-docs-server"]
}
}
}
Local Development
# Clone and setup
git clone <repository-url>
cd mcp-docs-server
npm install
# Build
npm run build
# Run locally
npm run dev
# Test with MCP Inspector
npx @modelcontextprotocol/inspector npm run dev
Usage 🚀
Getting Started
# Ask for getting started guide
mcp_docs_guide("getting_started")
# Quick start workflow
mcp_quick_start()
Finding Documentation
# Search for specific topics
search_docs("tools development")
search_docs("JSON-RPC protocol", "specification")
# Browse by category
get_docs_by_category("concepts")
get_docs_by_category("development")
Development Workflows
# Server development workflow
server_development_workflow("tool", "python")
# Client integration guide
client_integration_guide("desktop")
# Troubleshooting
troubleshooting_workflow("connection")
Accessing Documentation Resources
# Access specific documentation files
Resource: mcp-docs://docs-getting-started-intro.md
Resource: mcp-docs://docs-learn-architecture.md
Resource: mcp-docs://docs-develop-build-server.md
Documentation Categories 📖
The server organizes MCP documentation into these categories:
- 🚀 getting_started: Introduction and basic concepts
- 🧠 concepts: Architecture, primitives, and design principles
- 🛠️ development: Building servers and clients
- 📋 specification: Technical protocol details
- 🔧 tools: Development tools and debugging
- 👥 community: Governance and contribution guidelines
Example Workflows 💡
New Developer Workflow
- Start with
mcp_quick_start()for overview - Use
mcp_docs_guide("getting_started")for structured introduction - Follow
server_development_workflow()to build first server - Use
search_docs()to find specific implementation details
Experienced Developer Workflow
- Use
search_docs()to find specific information - Access documentation resources directly via URIs
- Reference
troubleshooting_workflow()when issues arise - Use
deployment_guide()for production deployment
Integration Developer Workflow
- Start with
client_integration_guide()for your platform - Use
get_docs_by_category("development")for implementation details - Reference protocol specification via
get_docs_by_category("specification") - Use debugging tools and workflows as needed
Architecture 🏗️
The server follows the same patterns as other MCP servers:
┌─────────────────────────────────┐
│ AI Application │
│ ┌─────────────────┐ │
│ │ MCP Client │ │
│ └─────────────────┘ │
└─────────────────────────────────┘
│
▼
┌─────────────────────────────────┐
│ MCP Docs Server │
│ │
│ 📚 Documentation Database │
│ 🔍 Search Engine │
│ 🎯 Guided Workflows │
│ 📄 Resource Access │
└─────────────────────────────────┘
Key Components
- Documentation Engine: Processes and categorizes scraped MCP docs
- Search System: Full-text search with relevance ranking
- Guide Generator: Creates structured learning workflows
- Resource Provider: Serves individual documentation files
- Prompt System: Interactive development workflows
Development 👨💻
Project Structure
mcp-docs-server/
├── src/
│ ├── index.ts # Main server entry point
│ ├── tools/ # MCP tools implementation
│ │ ├── mcpDocsGuide.ts # Main documentation guide
│ │ ├── searchDocs.ts # Documentation search
│ │ └── getDocsByCategory.ts # Category browsing
│ ├── prompts/ # Interactive workflows
│ │ ├── index.ts # Prompt registry
│ │ ├── quickStartGuide.ts
│ │ ├── serverDevelopmentWorkflow.ts
│ │ ├── clientIntegrationGuide.ts
│ │ ├── troubleshootingWorkflow.ts
│ │ └── deploymentGuide.ts
│ └── resources/ # Resource access
│ └── index.ts # Documentation file access
├── scraped_docs/ # MCP documentation files
├── package.json
├── tsconfig.json
└── README.md
Adding New Documentation
- Add markdown files to
scraped_docs/directory - Files are automatically categorized and indexed
- Search and resource access work immediately
- No server restart required
Contributing
- Fork the repository
- Create a feature branch
- Make your changes
- Add tests if applicable
- Submit a pull request
Troubleshooting 🔧
Server Not Starting
# Check TypeScript compilation
npm run build
# Test directly
node build/index.js
# Check with inspector
npx @modelcontextprotocol/inspector node build/index.js
Claude Desktop Integration Issues
- Check configuration syntax in
claude_desktop_config.json - Use absolute paths if needed
- Restart Claude Desktop completely
- Check Claude logs:
~/Library/Logs/Claude/mcp.log
Documentation Not Found
- Verify
scraped_docs/directory exists - Check file permissions
- Ensure markdown files are present
- Check server logs for errors
API Reference 📋
Tools
mcp_docs_guide
Get structured guides for MCP development topics.
Parameters:
topic(required): Topic to get guidance ongetting_started: Introduction and basicsbuilding_servers: Server development guidebuilding_clients: Client development guidecore_concepts: Architecture and primitivestools_and_resources: Deep dive into tools and resourcesprotocol_specification: Technical protocol detailstroubleshooting: Common issues and solutionsbest_practices: Development best practicesexamples_and_tutorials: Complete examples
search_docs
Search through MCP documentation using keywords.
Parameters:
query(required): Search query stringcategory(optional): Limit search to specific category
get_docs_by_category
Browse documentation organized by category.
Parameters:
category(required): Documentation category to explore
Resources
All documentation files are available as resources with URI format:
mcp-docs://filename.md
Example resources:
mcp-docs://docs-getting-started-intro.mdmcp-docs://docs-learn-architecture.mdmcp-docs://docs-develop-build-server.md
License 📄
MIT License - see LICENSE file for details.
Support 💬
- Issues: Report bugs and feature requests on GitHub
- Documentation: Use the server itself to explore MCP documentation
- Community: Join MCP community discussions
- Contributing: See CONTRIBUTING.md for guidelines
Built with ❤️ for the MCP community
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。