Cover image
Try Now
2025-04-14

GO-MCP是MCP SDK的强大GO(Golang)版本,它实现了模型上下文协议(MCP),以促进外部系统和AI应用程序之间的无缝通信。

3 years

Works with Finder

8

Github Watches

45

Github Forks

306

Github Stars

Go-MCP

Statusphere logo

Release Stars Forks Issues Pull Requests License Contributors Last Commit

Go Reference Go Report Go Tests

中文文档

🚀 Overview

Go-MCP is a powerful Go version of the MCP SDK that implements the Model Context Protocol (MCP) to facilitate seamless communication between external systems and AI applications. Based on the strong typing and performance advantages of the Go language, it provides a concise and idiomatic API to facilitate your integration of external systems into AI applications.

✨ Key Features

  • 🔄 Complete Protocol Implementation: Full implementation of the MCP specification, ensuring seamless integration with all compatible services
  • 🏗️ Elegant Architecture Design: Adopts a clear three-layer architecture, supports bidirectional communication, ensuring code modularity and extensibility
  • 🔌 Seamless Integration with Web Frameworks: Provides MCP protocol-compliant http.Handler, allowing developers to integrate MCP into their service frameworks
  • 🛡️ Type Safety: Leverages Go's strong type system for clear, highly maintainable code
  • 📦 Simple Deployment: Benefits from Go's static compilation, eliminating complex dependency management
  • High-Performance Design: Fully utilizes Go's concurrency capabilities, maintaining excellent performance and low resource overhead across various scenarios

🛠️ Installation

go get github.com/ThinkInAIXYZ/go-mcp

Requires Go 1.18 or higher.

🎯 Quick Start

Client Example

package main

import (
	"context"
	"log"

	"github.com/ThinkInAIXYZ/go-mcp/client"
	"github.com/ThinkInAIXYZ/go-mcp/transport"
)

func main() {
	// Create SSE transport client
	transportClient, err := transport.NewSSEClientTransport("http://127.0.0.1:8080/sse")
	if err != nil {
		log.Fatalf("Failed to create transport client: %v", err)
	}

	// Initialize MCP client
	mcpClient, err := client.NewClient(transportClient)
	if err != nil {
		log.Fatalf("Failed to create MCP client: %v", err)
	}
	defer mcpClient.Close()

	// Get available tools
	tools, err := mcpClient.ListTools(context.Background())
	if err != nil {
		log.Fatalf("Failed to list tools: %v", err)
	}
	log.Printf("Available tools: %+v", tools)
}

Server Example

package main

import (
	"fmt"
	"log"
	"time"

	"github.com/ThinkInAIXYZ/go-mcp/protocol"
	"github.com/ThinkInAIXYZ/go-mcp/server"
	"github.com/ThinkInAIXYZ/go-mcp/transport"
)

type TimeRequest struct {
	Timezone string `json:"timezone" description:"timezone" required:"true"` // Use field tag to describe input schema
}

func main() {
	// Create SSE transport server
	transportServer, err := transport.NewSSEServerTransport("127.0.0.1:8080")
	if err != nil {
		log.Fatalf("Failed to create transport server: %v", err)
	}

	// Initialize MCP server
	mcpServer, err := server.NewServer(transportServer)
	if err != nil {
		log.Fatalf("Failed to create MCP server: %v", err)
	}

	// Register time query tool
	tool, err := protocol.NewTool("current time", "Get current time for specified timezone", TimeRequest{})
	if err != nil {
		log.Fatalf("Failed to create tool: %v", err)
		return
	}
	mcpServer.RegisterTool(tool, handleTimeRequest)

	// Start server
	if err = mcpServer.Run(); err != nil {
		log.Fatalf("Server failed to start: %v", err)
	}
}

func handleTimeRequest(req *protocol.CallToolRequest) (*protocol.CallToolResult, error) {
	var timeReq TimeRequest
	if err := protocol.VerifyAndUnmarshal(req.RawArguments, &timeReq); err != nil {
		return nil, err
	}

	loc, err := time.LoadLocation(timeReq.Timezone)
	if err != nil {
		return nil, fmt.Errorf("invalid timezone: %v", err)
	}

	return &protocol.CallToolResult{
		Content: []protocol.Content{
			protocol.TextContent{
				Type: "text",
				Text: time.Now().In(loc).String(),
			},
		},
	}, nil
}

🏗️ Architecture Design

Go-MCP adopts an elegant three-layer architecture:

Architecture Overview

  1. Transport Layer: Handles underlying communication implementation, supporting multiple transport protocols
  2. Protocol Layer: Handles MCP protocol encoding/decoding and data structure definitions
  3. User Layer: Provides friendly client and server APIs

Currently supported transport methods:

Transport Methods

  • HTTP SSE/POST: HTTP-based server push and client requests, suitable for web scenarios
  • Stdio: Standard input/output stream-based, suitable for local inter-process communication

The transport layer uses a unified interface abstraction, making it simple to add new transport methods (like Streamable HTTP, WebSocket, gRPC) without affecting upper-layer code.

🤝 Contributing

We welcome all forms of contributions! Please see CONTRIBUTING.md for details.

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

📞 Contact Us

WeChat QR Code

✨ Contributors

Thanks to all developers who have contributed to this project!

Contributors

📈 Project Trends

Star History

相关推荐

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

  • 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

  • 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.

  • Beria Joey
  • 你的职业规划师,不走弯路就问我。Sponsor:小红书“ ItsJoe就出行 ”

  • 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

  • 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。

  • GLips
  • MCP服务器向像光标这样的AI编码代理提供FIGMA布局信息

    Reviews

    3 (1)
    Avatar
    user_M7dyWlqL
    2025-04-17

    I've been using go-mcp by ThinkInAIXYZ and it's fantastic! This application delivers a smooth and efficient performance consistent with MCP principles. It's easy to navigate and integrates seamlessly with my workflow. For anyone considering it, check out the official repository at https://github.com/ThinkInAIXYZ/go-mcp. Highly recommended!