Skip to main content
Glama
A2A MCP Server
by ww2521

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 agent

  • a2a_list_agents - 列出所有已发现的 agents

  • a2a_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

Related MCP server: MCPBridge

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

Available Tools

6 tools
a2a_cancel_taskC

Cancel a running task.

Inputs: agent (required), task_id (required).

a2a_discoverA

Discover an A2A agent at a URL. Fetches the agent card and registers it for later use.

Inputs: url (required).

a2a_get_taskB

Get the current status and result of a task by ID.

Inputs: agent (required), task_id (required).

a2a_list_agentsA

List all discovered A2A agents with their capabilities and skills.

Inputs: no named parameters.

a2a_list_tasksA

List tasks on a remote agent, optionally filtered by context.

Inputs: agent (required), page_size, context_id.

a2a_sendA

Send a message to an A2A agent and wait for the result. Supports polling for async tasks.

Inputs: agent (required), message (required), poll, context_id, +3 more.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.7/5.0

    Scored across 6 tools

    Disambiguation5/5

    Each tool has a clearly distinct purpose: discovery, listing agents, sending messages, and managing tasks by ID or listing. No overlap.

    Naming Consistency5/5

    All tools follow the consistent pattern 'a2a_verb_noun' in lowercase snake_case, with verbs like list, get, send, cancel, discover.

    Tool Count5/5

    With 6 tools covering discovery, task lifecycle (send, get, list, cancel), the count is well-scoped for an A2A protocol server.

    Completeness4/5

    Covers core A2A operations: agent discovery, task creation (via send), retrieval, listing, and cancellation. Minor gap: explicit task creation tool, but send likely handles it.

    Related MCP Connectors

    Related MCP Servers