MCP SpecNavigator
Enables intelligent navigation and exploration of the Model Context Protocol specification through dynamic markdown tree generation, section search, content retrieval, and upstream synchronization with the official MCP repository.
README
@lepion/mcp-server-specnavigator
An intelligent MCP (Model Context Protocol) server for navigating, exploring, and understanding the Model Context Protocol specification with dynamic markdown tree generation, intelligent section navigation, and upstream synchronization.
🎯 Purpose
The MCP SpecNavigator server transforms the way you interact with MCP specifications by providing:
- 🌳 Dynamic Markdown Trees - Navigate specifications with auto-generated tree structures
- 🔍 Intelligent Search - Find sections instantly with keyword search and anchor links
- 🔄 Upstream Sync - Stay updated with the latest specification changes
- 📚 Contextual Documentation - Access specific sections on-demand
- ⚡ Performance Optimized - Efficient caching for fast responses
📋 Attribution
This project is based on and provides navigation for the Model Context Protocol specification created by the Model Context Protocol team. This is a wrapper/navigation tool that enhances access to the original specification.
✨ Features
- Dynamic Markdown Tree Generation: Automatically generates navigable tree structures from MCP specifications
- Intelligent Section Search: Find sections by keywords and get direct anchor links
- Specification Caching: Efficiently caches specifications to reduce network requests
- Upstream Synchronization: Check for updates from the official MCP repository
- Section Content Retrieval: Get specific sections of the specification on demand
📦 Installation
Via npm (Recommended)
npm install -g @lepion/mcp-server-specnavigator
From Source
# Clone the repository
git clone https://github.com/kayaozkur/mcp-server-specnavigator.git
cd mcp-server-specnavigator
# Install dependencies
npm install
# Build the TypeScript code
npm run build
# Start the server
npm start
Development Mode
# Run with hot reload
npm run dev
🛠️ Available Tools
1. fetch_spec_updates
Check for upstream changes in the MCP specification repository.
Parameters:
force_refresh(boolean, optional): Force refresh even if cache is valid
Returns:
- Update status and any new changes detected
2. generate_markdown_tree
Create a navigable tree structure of the MCP specification.
Parameters:
max_depth(number, optional): Maximum depth of the tree (default: 3)include_anchors(boolean, optional): Include anchor links in the tree
Returns:
- Markdown-formatted tree structure with navigation links
3. find_section
Search for sections by keyword and return anchor links.
Parameters:
query(string, required): Search query for section titlesfuzzy(boolean, optional): Enable fuzzy matching (default: false)
Returns:
- Array of matching sections with titles, paths, and anchor links
4. get_spec_content
Retrieve content from a specific section of the specification.
Parameters:
section_path(string, required): Path to the section (e.g., "protocol/messages")include_subsections(boolean, optional): Include subsection content
Returns:
- Markdown content of the requested section
⚙️ MCP Configuration
For Claude Desktop
Add to your Claude Desktop configuration:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"specnavigator": {
"command": "npx",
"args": ["@lepion/mcp-server-specnavigator"],
"env": {
"NODE_ENV": "production"
}
}
}
}
For Development
{
"mcpServers": {
"specnavigator": {
"command": "node",
"args": ["/path/to/mcp-server-specnavigator/dist/index.js"],
"env": {
"NODE_ENV": "development"
}
}
}
}
🏗️ Architecture
src/
├── index.ts # Main MCP server entry point
├── spec-fetcher.ts # Handles fetching and caching specs
├── markdown-tree-generator.ts # Generates dynamic markdown trees
├── anchor-navigator.ts # Finds sections and generates anchors
└── tools.ts # MCP tool definitions
🧪 Development
Running Tests
npm test
Linting
npm run lint
Formatting
npm run format
🤝 Contributing
Contributions are welcome! Please feel free to submit a Pull Request. For major changes, please open an issue first to discuss what you would like to change.
📄 License
MIT License - see LICENSE file for details.
🙏 Acknowledgments
- Model Context Protocol Team - For creating the MCP specification that this server navigates
- Anthropic - For developing the Model Context Protocol standard
- Contributors - Everyone who has contributed to improving this navigation tool
🔗 Links
- NPM Package NOT AVAILABLE AT THIS POINT
- GitHub Repository
- Model Context Protocol Specification
- Report Issues
Built with ❤️ by the Lepion Team
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。