Mcpgo Ed1164e4

MCP运维云端

提供方 coohu

运维云端 未标注 会话登录 社区 / 第三方

一个高性能的Golang实现的MCP网关,提供反向代理和管理层功能,支持Kubernetes部署、多存储选项和全面的可观测性。

一句话选型结论

按资料字段整理

Mcpgo Ed1164e4 属于运维云端类 MCP 服务器,传输方式未标注,会话登录,没有现成安装命令,需按官方文档接入。接入前先确认权限范围;涉及写入、删除或付款的操作,建议先在测试环境验证。

分类运维云端 传输方式未标注 鉴权会话登录 部署方式未标注
GitHub 星标 0 仓库热度,无公开仓库时显示未标注
工具数 未标注 暴露给模型的工具数量
资源 / 提示词 0 / 0 资源与提示词模板数量
维护方 社区 / 第三方 核验等级 仅资料核对

工具与能力

运维云端
数量来自来源目录的标注,具体工具清单以官方文档为准
工具数量未标注暴露给模型可调用的工具个数
资源数量0可读取的资源条目
提示词数量0内置提示词模板数量

能力标签

适用场景

反向代理Go本地部署Kubernetes管理多存储支持可观测性高性能网关

安装与配置

未标注 会话登录
上游未给出传输方式
  1. 1
    准备运行环境

    按官方文档确认运行环境(常见为 Node.js 或 Python),装好后再执行安装命令。

  2. 2
    安装服务

    复制上面的安装命令执行;命令里带占位符的位置,换成自己的密钥或路径。

  3. 3
    写入客户端配置

    在客户端的 MCP 配置里填好启动命令和参数,保存后完全退出客户端再打开,让新配置生效。

  4. 4
    验证是否可用

    连接成功后随便调一个只读工具,确认能返回结果,再接入正式流程。

典型使用场景

3 类
按分类与标签整理的常见用法,具体以官方文档为准
云资源查询

查实例、存储、账单等云上资源状态。

部署与发布

把部署、重启、回滚这类操作接进对话。

监控与告警

拉取监控指标,异常时让模型帮你判断。

选择建议

选之前先看这几条
按条目字段给出的判断,不替你决定用不用

值得选的地方

  • 基础信息齐全,可以先按官方文档接入试跑。

先想清楚的地方

  • 没有现成安装命令,需要按官方文档自己填启动命令和参数。
  • 使用前要准备好会话登录对应的凭据,并按最小权限配置。
  • 资料没标注工具数量,实际能力要装好后确认。
  • 版本更新与行为细节以官方仓库为准。

更适合

  • 需要在运维云端场景里让模型直接调用外部能力的人。
  • 想先快速试一下、能接受在本地或自有环境跑服务的人。

不太适合

  • 不方便管理密钥或第三方授权的人。
  • 需要厂商级 SLA 和技术支持的生产环境。

同类对比

6 个同类
同分类里挑最接近的几条,字段以各自详情页为准
MCP 服务器工具数传输方式部署方式安装命令最后核验
Mcpgo Ed1164e4当前 未标注 未标注 未标注 无 2026-05-17
lobehub 未标注 stdio · 本地进程 本地部署 有 2026-04-11
TrendRadar 未标注 stdio · 本地进程 本地部署 有 2026-04-13
Ui Tars Desktop 8a825770 未标注 stdio · 本地进程 本地部署 有 2026-05-17
ruflo 未标注 stdio · 本地进程 本地部署 有 2026-04-13
FastGPT 未标注 stdio · 本地进程 本地部署 有 2026-04-11

服务器档案

运维云端
字段来自项目仓库与公开索引,中文说明由本站标注
提供方coohu上游标注的维护方 coohu
分类运维云端本站按上游分类展示
连接方式未标注上游未给出传输方式
鉴权方式会话登录远程服务建议用最小权限凭据
维护方社区 / 第三方核验等级 仅资料核对
包名 / 运行时未标注 · 未标注本地安装时用到的包标识与运行时
部署类型未标注上游给出的部署形态
最近核验2026-05-17上游最后核验时间

详细介绍

12.2 KB
来自项目仓库与上游目录的原文,未做删改

MCP Gateway (Golang)

A high-performance Golang implementation of the MCP Gateway - a reverse proxy and management layer for Model Context Protocol (MCP) servers.

✨ Features

  • ✅ Control Plane: RESTful APIs for MCP server and tool management
  • ✅ Data Plane: Session-aware routing and reverse proxy for MCP requests
  • ✅ CLI Tool: mcpctl for easy management, operations, and deployment
  • ✅ Multi-Storage: Support for in-memory, Redis, and Cosmos DB storage (in-memory & Redis complete)
  • ✅ Kubernetes Native: Deploy and manage MCP servers in Kubernetes with StatefulSets
  • ✅ Authentication & Authorization: Azure Entra ID integration with RBAC
  • ✅ Tool Gateway: Dynamic routing for registered MCP tools with MCP protocol support
  • ✅ Observability: OpenTelemetry tracing, Prometheus metrics, and structured logging
  • ✅ Multi-Architecture: Support for AMD64 and ARM64 platforms
  • ✅ Production Ready: Comprehensive error handling, graceful shutdown, health checks

📊 Status

Version: 1.0.0
Build Status: ✅ Passing
Test Coverage: 95%+
Progress: 95% Complete

Completed Components

  • ✅ Core data structures and contracts
  • ✅ In-memory and Redis storage
  • ✅ Kubernetes deployment manager
  • ✅ Session-aware routing with pod watching
  • ✅ HTTP reverse proxy with auth
  • ✅ Tool gateway with MCP protocol
  • ✅ OpenTelemetry observability stack
  • ✅ Comprehensive tests
  • ✅ Production deployment manifests

Architecture```

┌─────────────────────────────────────────────────────────────┐
│ MCP Gateway │
│ │
│ ┌────────────────┐ ┌─────────────────┐ │
│ │ Control Plane │ │ Data Plane │ │
│ │ │ │ │ │
│ │ /adapters │ │ /adapters/:id │ │
│ │ /tools │ │ /mcp │ │
│ └────────────────┘ └─────────────────┘ │
│ │ │ │
│ └─────────┬───────────────────┘ │
│ │ │
│ ┌───────────▼──────────┐ │
│ │ Deployment Manager │ │
│ └───────────┬──────────┘ │
└────────────────────┼────────────────────────────────────────┘
│
┌───────────▼────────────┐
│ Kubernetes Cluster │
│ │
│ ┌──────┐ ┌──────┐ │
│ │ MCP │ │ Tool │ │
│ │Server│ │Server│ │
│ └──────┘ └──────┘ │
└────────────────────────┘


## Project Structure```
mcpgo/
├── cmd/
│   ├── gateway/         # Gateway service entry point
│   ├── toolrouter/      # Tool router service entry point
│   └── mcpctl/          # CLI client for management and operations
├── internal/
│   ├── api/             # HTTP API controllers
│   ├── auth/            # Authentication & authorization
│   ├── cli/             # CLI commands and client
│   ├── config/          # Configuration management
│   ├── contracts/       # Data models and interfaces
│   ├── deployment/      # Kubernetes deployment manager
│   ├── middleware/      # HTTP middleware
│   ├── observability/   # OpenTelemetry tracing, metrics, logging
│   ├── proxy/           # HTTP reverse proxy
│   ├── router/          # Session routing & K8s node provider
│   ├── service/         # Business logic services
│   ├── session/         # Session management
│   ├── store/           # Persistence layer (memory, Redis, Cosmos)
│   └── toolgateway/     # Tool gateway router
├── docs/                # Documentation
│   ├── CLI.md           # CLI documentation
│   ├── QUICKSTART.md    # Quick start guide
│   ├── DEPLOYMENT.md    # Deployment guide
│   ├── observability.md # Observability guide
│   └── KUBERNETES_DEPLOYMENT.md  # K8s details
├── examples/            # Example configurations
│   ├── adapter-example.json
│   └── tool-example.json

Quick Start

Prerequisites

  • Go 1.21 or later
  • Docker
  • Kubernetes (local or cloud)
  • Redis (for development)

Build```bash

Download dependencies

make deps

Build binaries

make build

Run tests

make test

Run linter

make lint


### Run Locally```bash
# Run gateway service
make run-gateway

# Or with custom config
./bin/gateway -config config.example.yaml

# Run tool router
./bin/toolrouter -config config.example.yaml

CLI Management Tool

Build and use the CLI tool for easy management:```bash

Build CLI

make build

Install system-wide (optional)

make install-mcpctl

Initialize CLI configuration

./bin/mcpctl config init --api-url http://localhost:8080

Create an adapter

./bin/mcpctl adapter create -f examples/adapter-example.json

List adapters

./bin/mcpctl adapter list

Scale an adapter

./bin/mcpctl deploy scale adapter my-adapter 5

Check gateway health

./bin/mcpctl ops health


See CLI Documentation and Quick Start Guide for more details.

### Docker```bash
# Build Docker images
make docker-build VERSION=1.0.0

# Push to registry
make docker-push DOCKER_REGISTRY=your-registry VERSION=1.0.0

# Build multi-architecture images
make docker-build-multiarch

Deploy to Kubernetes```bash

Deploy all resources

make k8s-deploy

Port forward for local access

kubectl port-forward -n adapter svc/mcp-gateway 8000:8000


## Configuration

The gateway can be configured via:
1. Configuration file (YAML)
2. Environment variables (prefixed with `MCPG_`)
3. Command-line flags

Example configuration:```yaml
server:
  port: 8000
  environment: development
  shutdown_timeout: 30

redis:
  enabled: true
  connection_string: localhost:6379

kubernetes:
  namespace: adapter
  in_cluster: false
  kubeconfig_path: ~/.kube/config

auth:
  enabled: false
  admin_role: mcp.admin

registry:
  endpoint: docker.io

observability:
  tracing_enabled: true
  tracing_collector: localhost:4317
  tracing_sample_rate: 1.0
  metrics_enabled: true
  metrics_collector: localhost:4317
  log_level: info

API Endpoints

Gateway Service

Control Plane - Adapter Management

  • POST /adapters - Create MCP adapter
  • GET /adapters - List MCP adapters
  • GET /adapters/:name - Get adapter details
  • PUT /adapters/:name - Update adapter
  • DELETE /adapters/:name - Delete adapter
  • GET /adapters/:name/status - Get deployment status
  • GET /adapters/:name/logs?instance=0 - Get pod logs

Control Plane - Tool Management

  • POST /tools - Register tool
  • GET /tools - List tools
  • GET /tools/:name - Get tool details
  • PUT /tools/:name - Update tool
  • DELETE /tools/:name - Delete tool
  • GET /tools/:name/status - Get tool status
  • GET /tools/:name/logs?instance=0 - Get tool logs

Data Plane - Proxy Endpoints

  • POST /adapters/:name/mcp - Proxy to specific MCP adapter
  • POST /mcp - Proxy to tool gateway router
  • GET /ping - Health check

Tool Router Service

MCP Protocol Endpoints

  • POST /tools/list - List available tools (MCP protocol)
  • POST /tools/call - Execute tool (MCP protocol)
  • GET /ping - Health check

Development

Install Development Tools```bash

make install-tools


### Code Formatting```bash
make fmt

Generate Mocks```bash

make gen-mocks


### Run Benchmarks```bash
make benchmark

Testing```bash

Run all tests

make test

Run tests with coverage

make test-coverage

Run specific test

go test -v ./internal/service/...


## Deployment

### Local Development

1. Start Redis: `docker run -d -p 6379:6379 redis:7`
2. Start gateway: `make run-gateway`
3. Test API: `curl http://localhost:8000/ping`

### Kubernetes

See docs/DEPLOYMENT.md for comprehensive deployment guide including:
- Local development setup
- Kubernetes production deployment
- OpenTelemetry Collector configuration
- High availability setup
- Security best practices

## Observability

The gateway includes comprehensive observability features powered by OpenTelemetry.

### Distributed Tracing

OpenTelemetry-based distributed tracing with OTLP exporter support:```yaml
observability:
  tracing_enabled: true
  tracing_collector: localhost:4317
  tracing_sample_rate: 1.0  # 100% sampling for dev, lower in prod

All HTTP requests are automatically traced with context propagation.

Metrics

Rich metrics collection for monitoring:

  • HTTP Metrics: requests_total, request_duration, requests_in_flight
  • Proxy Metrics: proxy_requests_total, proxy_duration, proxy_errors
  • Tool Metrics: tool_executions_total, tool_duration, tool_errors
  • Session Metrics: sessions_active, sessions_total
  • K8s Metrics: pods_healthy, pods_unhealthy

Metrics are exported via OTLP to Prometheus or other backends.

Structured Logging

Zap-based structured logging with:
- JSON format in production
- Colorized console logs in development
- Automatic request ID correlation
- Stack traces for errorsyaml observability: log_level: info # debug, info, warn, error

See docs/observability.md for detailed configuration.

Performance

Performance characteristics compared to .NET baseline:

Metric Go .NET Improvement
Binary Size ~68MB ~180MB 62% smaller
Memory (idle) ~50MB ~150MB 67% less
Memory (load) ~100MB ~300MB 67% less
Cold Start <1s ~5s 5x faster
Request Latency (p99) <5ms ~10ms 2x faster
Throughput >10k req/s ~5k req/s 2x higher
  • Lightweight footprint
  • Fast request processing
  • Horizontal scaling with session affinity
  • Efficient connection pooling and HTTP/2 support
  • Zero-allocation JSON encoding for logs

Comparison with .NET Version

Feature .NET Go Status
Adapter Management ✅ ✅ Complete
Tool Management ✅ ✅ Complete
K8s Deployment ✅ ✅ Complete
Session Routing ✅ ✅ Complete
HTTP Proxy ✅ ✅ Complete
Tool Gateway ✅ ✅ Complete
Auth & RBAC ✅ ✅ Complete
Observability ✅ (AppInsights) ✅ (OpenTelemetry) Complete
In-Memory Store ✅ ✅ Complete
Redis Store ✅ ✅ Complete
Cosmos DB Store ✅ ⏳ Pending
Multi-Architecture ❌ ✅ Go advantage
Container Size Large Small Go advantage
Startup Time Slow Fast Go advantage

Contributing

  1. Fork the repository
  2. Create a feature branch
  3. Make your changes
  4. Run tests and linter
  5. Submit a pull request

License

MIT License - see LICENSE for details

Documentation

  • 📖 Deployment Guide - Complete deployment instructions
  • 📊 Observability Guide - Tracing, metrics, and logging
  • ⚙️ Kubernetes Deployment - K8s technical details
  • 📝 Project Status - Implementation status and roadmap
  • 🔧 Configuration Example - Sample configuration

Support

  • 📖 Documentation
  • 🐛 Issue Tracker
  • 💬 Discussions

Acknowledgments

This project is a Go implementation of the [MCP Gateway]( reference architecture, providing improved performance and reduced resource consumption.

常见问题

5 问
答案由本站依据条目字段整理,不引用上游原文
Mcpgo Ed1164e4 怎么安装?

资料没有给出安装命令,请到官方仓库或文档按说明接入;这类条目多数是生态应用,或尚未发布可安装包。

Mcpgo Ed1164e4 怎么在客户端里配置?

它的传输方式未标注,在客户端 MCP 配置里按官方文档填写启动命令和参数,保存后重启客户端生效。

Mcpgo Ed1164e4 收费吗?

本站不代收费用,也没有标注具体价格。是否收费取决于服务本身和背后的接口额度,正式使用前按官方说明确认。

Mcpgo Ed1164e4 需要 API Key 吗?

资料标注需要会话登录认证,使用前准备好对应凭据,并按最小权限配置。

Mcpgo Ed1164e4 能提供哪些能力?

资料显示能力标签包括 Go、可观测性、云端部署。具体工具清单以官方文档为准。

热门专题

按标签挑服务器
全部专题 →

服务器信息以作者仓库与官方文档为准;分类、传输方式与鉴权方式的中文说明为中文整理,最近核验 2026-05-17。