
mcp4go
A comprehensive Go SDK for the Model Context Protocol (MCP) - simplifying AI application development by abstracting away protocol complexities.
Repository Info
About This Server
A comprehensive Go SDK for the Model Context Protocol (MCP) - simplifying AI application development by abstracting away protocol complexities.
Model Context Protocol (MCP) - This server can be integrated with AI applications to provide additional context and capabilities, enabling enhanced AI interactions and functionality.
Documentation
MCP4Go
MCP4Go is a Go implementation of the Model Context Protocol (MCP), designed to simplify the development of AI applications by abstracting away protocol complexities.
Features
- Complete MCP protocol implementation in pure Go
- High-level abstractions for common MCP resources
- Pluggable architecture for custom extensions
- Comprehensive documentation and examples
- Production-ready with robust error handling
Installation
MCP4Go requires Go 1.18 or later. Install it using Go modules:
go get github.com/mcp4go/mcp4go
Getting Started
To get started with MCP4Go, import the package in your Go application:
package main
import (
"context"
"fmt"
"log"
"os"
"os/signal"
"syscall"
"time"
"github.com/mcp4go/mcp4go/pkg/logger"
"github.com/mcp4go/mcp4go/protocol"
"github.com/mcp4go/mcp4go/server"
"github.com/mcp4go/mcp4go/server/iface"
"github.com/mcp4go/mcp4go/server/transport"
_ "time/tzdata" // Load all time zones
)
func main() {
ctx, cancel := context.WithCancel(context.Background())
defer cancel()
_logger := logger.DefaultLog
// Listen for interrupt signals for graceful exit
signalCh := make(chan os.Signal, 1)
signal.Notify(signalCh, os.Interrupt, syscall.SIGTERM)
go func() {
<-signalCh
_logger.Logf(ctx, logger.LevelWarn, "Received shutdown signal")
cancel()
}()
type timeRequest struct {
TimeZone string `json:"time_zone,omitempty" description:"time zone default is Asia/Shanghai"`
}
// Create standard input/output transport layer
stdioTransport := transport.NewStdioTransport()
// Create MCP server and configure options
srv, cleanup, err := server.NewServer(
stdioTransport,
server.WithServerInfo(protocol.Implementation{
Name: "time-mcp",
Version: "0.1.0",
}),
server.WithInstructions("Welcome to Time MCP! This server provides Time tools."),
server.WithToolBuilder(iface.NewFunctionalToolsBuilder(
iface.NewFunctionalToolWrapper(
"time",
"get current time",
func(ctx context.Context, args timeRequest) ([]protocol.Content, error) {
if args.TimeZone == "" {
args.TimeZone = "Asia/Shanghai"
}
getTimezoneTime := func(_ context.Context, timeZone string) time.Time {
if timeZone == "" {
timeZone = "Asia/Shanghai"
}
loc, err := time.LoadLocation(timeZone)
if err != nil {
return time.Now()
}
return time.Now().In(loc)
}
currentTime := getTimezoneTime(ctx, args.TimeZone)
return []protocol.Content{
protocol.NewTextContent(fmt.Sprintf("current time is %s", currentTime), nil),
}, nil
}),
)),
)
if err != nil {
log.Printf("Failed to create server: %v\n", err)
return
}
defer cleanup()
// Start the server
_logger.Logf(ctx, logger.LevelWarn, "Starting Time MCP server")
if err := srv.Run(ctx); err != nil {
_logger.Logf(ctx, logger.LevelError, "Server error: %v", err)
return
}
_logger.Logf(ctx, logger.LevelWarn, "Server shutdown complete")
}
Protocol Implementation
MCP4Go provides a complete implementation of the Model Context Protocol with support for:
- JSON-RPC communication
- Resource lifecycle management
- Prompt engineering
- Tool definitions and invocations
- Sampling parameters
- Logging and diagnostics
License
This project is licensed under the MIT License
Contributing
Contributions are welcome! Please see our Contributing Guide for more information.
Quick Start
Clone the repository
git clone https://github.com/mcp4go/mcp4goInstall dependencies
cd mcp4go
npm installFollow the documentation
Check the repository's README.md file for specific installation and usage instructions.
Repository Details
Recommended MCP Servers
Discord MCP
Enable AI assistants to seamlessly interact with Discord servers, channels, and messages.
Knit MCP
Connect AI agents to 200+ SaaS applications and automate workflows.
Apify MCP Server
Deploy and interact with Apify actors for web scraping and data extraction.
BrowserStack MCP
BrowserStack MCP Server for automated testing across multiple browsers.
Zapier MCP
A Zapier server that provides automation capabilities for various apps.