A2A MCP Server
Exposes A2A (Agent-to-Agent) protocol as MCP tools, enabling Goose to call remote A2A agents via External Subagent mechanism.
README
A2A MCP Server
将 A2A (Agent-to-Agent) 协议暴露为 MCP tools,让 Goose 可以通过 External Subagent 机制调用远程 A2A agents。
功能
a2a_discover- 发现并注册远程 A2A agenta2a_list_agents- 列出所有已发现的 agentsa2a_send- 发送消息给 agent 并等待结果(支持轮询)a2a_get_task- 查询任务状态a2a_cancel_task- 取消运行中的任务a2a_list_tasks- 列出 agent 上的任务
环境变量
| 变量 | 说明 | 默认值 |
|---|---|---|
A2A_GATEWAY_URL |
A2A Gateway URL (如 LiteLLM) | http://localhost:4000 |
A2A_API_KEY |
认证 API Key | 空 |
A2A_TIMEOUT |
HTTP 请求超时 (毫秒) | 30000 |
安装
cd a2a-mcp-server
npm install
npm run build
配置 Goose External Subagent
在 ~/.config/goose/config.yaml 中添加:
extensions:
a2a-mcp-server:
type: stdio
cmd: node
args:
- /path/to/a2a-mcp-server/dist/server.js
env:
A2A_GATEWAY_URL: "http://localhost:4000"
A2A_API_KEY: "your-api-key"
timeout: 300
bundled: false
enabled: true
或者使用绝对路径的 wrapper script:
#!/bin/bash
# /usr/local/bin/a2a-mcp-server
export A2A_GATEWAY_URL="http://localhost:4000"
export A2A_API_KEY="your-api-key"
exec node /path/to/a2a-mcp-server/dist/server.js
然后配置:
extensions:
a2a-mcp-server:
type: stdio
cmd: /usr/local/bin/a2a-mcp-server
timeout: 300
bundled: false
enabled: true
使用示例
1. 发现 Agent
使用 a2a_discover 发现 http://localhost:10001 的 agent
2. 发送消息
用 a2a_send 给 my-agent 发送消息:"帮我分析这段代码的性能问题"
3. 查询任务
用 a2a_get_task 查询任务 task-12345 的状态
4. 列出所有 agents
用 a2a_list_agents 看看有哪些可用的 agents
架构
Goose Main Agent (GUI)
│
├── LLM: Claude/GPT/etc.
│
├── External Subagent: a2a-mcp-server (stdio MCP)
│ │
│ └── A2A Client
│ │
│ └── HTTP/JSON-RPC (A2A Protocol)
│ │
│ ▼
│ LiteLLM Gateway
│ │
│ ┌───────┼───────┐
│ ▼ ▼ ▼
│ Agent1 Agent2 Agent3
│
└── 其他 extensions...
与 LiteLLM 集成
如果你使用 LiteLLM 作为 A2A Gateway:
# 启动 LiteLLM
litellm --config config.yaml
# 配置 A2A MCP Server
export A2A_GATEWAY_URL="http://localhost:4000"
export A2A_API_KEY="sk-xxx"
LiteLLM 会自动代理 A2A 请求到上游 agents,并提供:
- 认证管理
- 请求日志
- 花费追踪
- 负载均衡
开发
# 开发模式
npm run dev
# 编译
npm run build
# 运行
npm start
License
MIT
Recommended Servers
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.
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.
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.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
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.
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.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
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.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.