kimi-bridge

kimi-bridge

An MCP server that enables WorkBuddy to invoke the kimi-code CLI locally via MCP protocol, allowing remote task execution through QQ messages.

Category
Visit Server

README

WorkBuddy Kimi Bridge

让 WorkBuddy 通过 MCP 协议 调用本地 kimi-code CLI,实现远程通过 QQ 消息控制本地 AI 执行任务。

架构:bridge-core (gRPC) → bridge-api (HTTP) → bridge-mcp-server (MCP) → WorkBuddy


✨ 功能特性

  • 三层架构:gRPC 核心调度 + HTTP API 层 + MCP 协议适配
  • 真实调度器:直接 spawn kimi-code 子进程,捕获 stdout 输出
  • MCP 原生接入:WorkBuddy 自动发现 7 个工具,零配置调用
  • 事件流驱动:全局事件队列,任务状态实时同步
  • 跨平台:支持 Windows(自动检测 Git Bash)

🏗️ 架构

┌─────────────┐     QQ 消息      ┌──────────────┐
│   用户手机   │ ───────────────→ │  WorkBuddy   │
└─────────────┘                  └──────┬───────┘
                                        │ MCP Streamable HTTP
                                        ↓
                              ┌─────────────────────┐
                              │ bridge-mcp-server   │  port 9001
                              │  (MCP 协议适配)      │
                              └──────────┬──────────┘
                                         │ HTTP REST
                                         ↓
                              ┌─────────────────────┐
                              │    bridge-api       │  port 8742
                              │   (HTTP API 层)     │
                              └──────────┬──────────┘
                                         │ gRPC
                                         ↓
                              ┌─────────────────────┐
                              │    bridge-core      │  port 50051
                              │  (进程调度核心)      │
                              └──────────┬──────────┘
                                         │ subprocess
                                         ↓
                              ┌─────────────────────┐
                              │   kimi-code CLI     │
                              │  (本地 AI 执行)      │
                              └─────────────────────┘

🚀 快速开始

前置要求

  • Python 3.12+
  • kimi-code CLI 已安装并登录
  • Git for Windows(bridge-core 自动检测路径)

安装依赖

pip install -r requirements.txt

启动服务

# PowerShell
Set-ExecutionPolicy -ExecutionPolicy RemoteSigned -Scope Process
.\start-all.ps1

三个服务将依次启动:

服务 端口 说明
bridge-core 50051 gRPC 核心调度
bridge-api 8742 HTTP API
bridge-mcp-server 9001 MCP 服务器

WorkBuddy 配置

编辑文件:

C:\Users\<你的用户名>\.workbuddy\connectors\<connector-id>\mcp.json

添加配置:

{
  "mcpServers": {
    "kimi-bridge": {
      "type": "streamable-http",
      "url": "http://127.0.0.1:9001/mcp",
      "timeout": 300000
    }
  }
}

注意:修改后需重启 WorkBuddy 生效。

💬 使用示例

在 QQ 给 WorkBuddy 发消息:

"用 kimi 帮我写一个 Python 快速排序,保存到 C:/Users/lenovo/Documents/quicksort.py"

WorkBuddy 会自动:

  1. 识别需求,调用 MCP run_task
  2. bridge 创建任务 → 启动 kimi 子进程
  3. kimi 执行代码生成和文件写入
  4. 返回结果到 QQ

⚠️ 重要提示

文件操作必须使用绝对路径,因为 kimi-code CLI 不会自动使用子进程的工作目录。

✅ 正确:C:/Users/lenovo/Documents/project/hello.py
❌ 错误:./hello.py

📁 项目结构

.
├── bridge-core/              # gRPC 核心调度器
│   ├── server.py             # 真实子进程调度(spawn kimi CLI)
│   ├── proto/bridge.proto    # gRPC 协议定义
│   └── generated/            # 生成的 pb 文件
├── bridge_api/               # HTTP API 层(FastAPI)
│   ├── main.py               # FastAPI 应用入口
│   ├── api/                  # REST 路由
│   ├── services/             # 业务逻辑(task_service, event_processor)
│   ├── storage/              # SessionStore(内存 + 持久化)
│   └── core_client/          # gRPC 客户端
├── bridge-mcp-server.py      # MCP 服务器(FastMCP)
├── bridge-gateway/           # Next.js Web 网关(可选)
├── tests/                    # pytest 测试套件
├── dev_files/                # 设计文档与交付总结
├── docs/                     # 使用说明与配置指南
├── start-all.ps1             # 一键启动脚本
└── requirements.txt          # Python 依赖

🛠️ MCP 工具列表

工具 说明
create_task 创建任务,返回 session_id
get_task_status 查询任务状态
get_task_logs 查询任务输出日志
list_tasks 列出所有任务
stop_task 停止任务
approve_task 审批等待中的工具调用
run_task 创建任务并阻塞轮询直到完成(最常用)

📄 文档

📜 License

MIT

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured