Dynamic MCP Server Framework
@scitara-cto
关于 Dynamic MCP Server Framework
A flexible and extensible framework for building Model Context Protocol (MCP) servers that conforms to the [Model Context Protocol specification](
基本信息
配置
工具
未检测到工具
工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。
概览
What is Dynamic MCP Server Framework?
Dynamic MCP Server Framework is a flexible, extensible framework for building Model Context Protocol (MCP) servers. It supports API key–based authentication, MongoDB-backed user management and authorization, session-based per-user tool loading, and dynamic tool sharing with fine-grained access control.
How to use Dynamic MCP Server Framework?
Install via npm or clone the repository, then follow the Getting Started guide. Tools are registered as handler packages; the server is configured programmatically. Clients authenticate by passing the user’s apiKey as a query parameter (e.g., /sse?apiKey=...).
Key features of Dynamic MCP Server Framework
- Dynamic tool management: create, delete, and authorize tools at runtime
- User management: add, update, delete, and list users with role-based access
- Tool sharing: share tools with other users and manage access levels
- Modern authentication: simple API key auth with MongoDB for authorization
- Extensibility: add custom HTTP routes and MongoDB collections in downstream projects
- Session-based tool loading: tools are loaded per user session, not globally
Use cases of Dynamic MCP Server Framework
- Building secure, user-aware AI tool servers that require per-user authentication
- Enabling runtime extensibility where tools can be added or modified without restarting
- Implementing multi-tenant environments where each user or session has tailored tool access
- Creating collaborative tool ecosystems where tools can be shared and permissions managed
FAQ from Dynamic MCP Server Framework
What authentication method does the server use?
The server uses API key authentication. Each user is assigned a unique apiKey, which clients provide as a query parameter (e.g., /sse?apiKey=...) to authenticate.
How are tools managed and registered?
Tools are registered via handler packages. Each package groups related tools and provides a handler function for execution. Tools can be created, deleted, and authorized at runtime—not limited to static startup definitions.
Can tools be shared with other users?
Yes. The framework supports tool sharing with fine-grained access control. Users can share tools with others and manage access levels through provided API actions.
Is the server extensible for downstream projects?
Yes. The framework allows adding custom HTTP routes and MongoDB collections, enabling downstream projects to extend the server’s functionality.
What is the tool visibility model?
All tools are visible to users by default. Users can hide tools from their view using the hide-tool and unhide-tool actions.
开发工具 分类下的更多 MCP 服务器
mcp-excalidraw
yctimlinMCP server and Claude Code skill for Excalidraw — programmatic canvas toolkit to create, edit, and export diagrams via AI agents with real-time canvas sync.
MCP-Scan: An MCP Security Scanner
invariantlabs-aiSecurity scanner for AI agents, MCP servers and agent skills.
🔐 ssh-mcp-server
classfang基于 SSH 的 MCP 服务 🧙♀️。已被MCP官方收录 🎉。 SSH MCP Server 🧙♀️. It has been included in the community MCP repository 🎉.
Golf
golf-mcpProduction-Ready MCP Server Framework • Build, deploy & scale secure AI agent infrastructure • Includes Auth, Observability, Debugger, Telemetry & Runtime • Run real-world MCPs powering AI Agents
JetBrains MCP Proxy Server
JetBrainsA model context protocol server to work with JetBrains IDEs: IntelliJ, PyCharm, WebStorm, etc. Also, works with Android Studio
评论