Cover image

功能强大的VSCODE扩展程序,可在Copilot中使用MCP服务器,从而访问MCP工具,资源等。

3 years

Works with Finder

4

Github Watches

17

Github Forks

303

Github Stars

Copilot MCP Client for VSCode

Please Read!

Since the VSCode team has offically brought MCP support to Copilot, this extension will now defer to their implementation. All existing configurations will be automatically migrated to the official VSCode implementation.

image

Version License: GPL v3 VSCode Extension MCP Client

A powerful VSCode extension that acts as a Model Context Protocol (MCP) client, enabling seamless integration between MCP tool servers and GitHub Copilot Chat. Join the growing ecosystem of interoperable AI applications with flexible integration options.

📑 Table of Contents

✨ Features

  • 🔧 MCP Server Management: Connect and manage multiple MCP servers through an intuitive UI
  • 🚀 Copilot Integration: Expose MCP tools directly to GitHub Copilot Chat participants
  • 🎯 Tool Discovery: Automatically discover and surface available tools from connected MCP servers
  • Server Health Monitoring: Real-time monitoring of MCP server status and connections
  • 🔄 Automatic Connection Management: Seamless handling of MCP server connections and reconnections
  • 🛠️ Tool Invocation Support: Full support for MCP tool invocation through Copilot Chat
  • 🔄 Multiple Server Types: Support for both process-based and SSE (Server-Sent Events) servers
  • 🛡️ Configuration Migration: Automatic migration of server configurations to the latest format
  • 🧩 Server Configuration Command: Manual migration command for updating older configurations

🚀 Usage

  1. Open the MCP Servers view from the VSCode activity bar
  2. Add and configure your MCP servers
  3. Enable/disable servers as needed
  4. Use GitHub Copilot Chat with your connected MCP tools using the @mcp participant
  5. View server status and tool availability in real-time

❓ FAQ

Why aren't my servers appearing in the server list?

This is typically caused by connection timeouts between the MCP SDK and your servers. Here's how to fix it:

  1. Use absolute paths for your executables - Instead of relative paths, specify the full path to your executable:

    /home/<user>/.nvm/versions/node/v<version>/bin/npx figma-developer-mcp --stdio --figma-api-key=...
    
  2. Find your executable paths - Not sure where your executables are located? Run this in your terminal:

    which npx    # or node, uvx, etc.
    

Using absolute paths helps prevent connection issues and ensures the MCP SDK can reliably launch your servers.

Why doesn't tool calling work with Claude 3.7?

The tool calling limitations are imposed by the VSCode API and Copilot Chat. Unfortunately, the tool calling models are limited to Claude 3.5 and GPT-4o. If you still face issues with tool calls, try switching to one or the other just for the tool you want to run. You can force a tool call in these models by using the '#' key and selecting the tool manually in your query.

How can I use the MCP server tools in Copilot Edits?

You can use the tools in Copilot Edits mode by manually specifying the tool(s) with the '#' key, and selecting the tool from the list:

Using tools in Copilot Edits with # key

Why isn't Copilot calling my tool?

GitHub Copilot has its own internal system prompt that may be dissuading tool calls in the chat. If you find that Copilot isn't using your tools automatically, try using the manual tool selection method mentioned above.

🔗 Requirements

  • VSCode
  • GitHub Copilot Chat extension
  • Compatible MCP servers (see Example Servers)

🌟 Benefits

  • Enable Copilot to use custom context and tools through MCP
  • Join the growing ecosystem of interoperable AI applications
  • Support local-first AI workflows
  • Flexible integration options for your development workflow

👥 Contributing

Contributions, issues and feature requests are welcome! Feel free to check the issues page.

✍️ Author

Vikash Loomba

📝 License

Copyright © 2024 Vikash Loomba.

This project is licensed under the GNU General Public License v3.0.


Part of the MCP Client Ecosystem - Enabling interoperable AI tools for developers ⭐️

相关推荐

  • Joshua Armstrong
  • Confidential guide on numerology and astrology, based of GG33 Public information

  • https://suefel.com
  • Latest advice and best practices for custom GPT development.

  • Emmet Halm
  • Converts Figma frames into front-end code for various mobile frameworks.

  • Elijah Ng Shi Yi
  • Advanced software engineer GPT that excels through nailing the basics.

  • https://maiplestudio.com
  • Find Exhibitors, Speakers and more

  • Alexandru Strujac
  • Efficient thumbnail creator for YouTube videos

  • Yusuf Emre Yeşilyurt
  • I find academic articles and books for research and literature reviews.

  • https://zenepic.net
  • Embark on a thrilling diplomatic quest across a galaxy on the brink of war. Navigate complex politics and alien cultures to forge peace and avert catastrophe in this immersive interstellar adventure.

  • https://reddgr.com
  • Delivers concise Python code and interprets non-English comments

  • 林乔安妮
  • A fashion stylist GPT offering outfit suggestions for various scenarios.

  • pontusab
  • 光标与风浪冲浪社区,查找规则和MCP

  • av
  • 毫不费力地使用一个命令运行LLM后端,API,前端和服务。

  • GeyserMC
  • 与Minecraft客户端/服务器通信的库。

  • Mintplex-Labs
  • 带有内置抹布,AI代理,无代理构建器,MCP兼容性等的多合一桌面和Docker AI应用程序。

  • 1Panel-dev
  • 🔥1Panel提供了直观的Web接口和MCP服务器,用于在Linux服务器上管理网站,文件,容器,数据库和LLMS。

  • awslabs
  • AWS MCP服务器 - 将AWS最佳实践直接带入您的开发工作流程的专门MCP服务器

  • 1Panel-dev
  • 💬MAXKB是一种现成的AI聊天机器人,它集成了检索功能增强的生成(RAG)管道,支持强大的工作流程并提供高级的MCP工具使用功能。

  • WangRongsheng
  • 🧑‍🚀 llm 资料总结(数据处理、模型训练、模型部署、 o1 模型、mcp 、小语言模型、视觉语言模型)|摘要世界上最好的LLM资源。

  • appcypher
  • 很棒的MCP服务器 - 模型上下文协议服务器的策划列表

  • chongdashu
  • 使用模型上下文协议(MCP),启用Cursor,Windsurf和Claude Desktop等AI助手客户,以通过自然语言控制虚幻引擎。

    Reviews

    4 (1)
    Avatar
    user_SLiJiIi8
    2025-04-17

    As a loyal user of copilot-mcp, I am thoroughly impressed with its capabilities. It offers seamless integration and elevates productivity tremendously. VikashLoomba has done a fantastic job in developing this tool. The interface is user-friendly, and the welcome information sets a positive tone from the start. Highly recommended for anyone looking to optimize their workflow! Check it out at https://github.com/VikashLoomba/copilot-mcp.