A2A-MCP-Server
If you are the rightful owner of A2A-MCP-Server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.
A2A MCP Server bridges the Model Context Protocol (MCP) with the Agent-to-Agent (A2A) protocol, enabling seamless interaction between MCP-compatible AI assistants and A2A agents.
The A2A MCP Server acts as an integration layer between two advanced AI agent protocols: the Model Context Protocol (MCP) and the Agent-to-Agent Protocol (A2A). MCP, developed by Anthropic, standardizes the connection of AI applications and large language models to external resources securely and composably. A2A, developed by Google, facilitates communication and interoperability between different AI agents through a standardized JSON-RPC interface. By bridging these protocols, the server allows MCP clients, such as Claude, to discover, register, communicate with, and manage tasks on A2A agents through a unified interface. This integration enhances the capabilities of AI assistants by enabling them to leverage the functionalities of various A2A agents, thereby expanding their utility and effectiveness in handling complex tasks.
Features
- Agent Management: Register, list, and unregister A2A agents with the bridge server.
- Communication: Send messages to A2A agents and receive responses, with real-time streaming support.
- Task Management: Track tasks handled by A2A agents, retrieve results using task IDs, and cancel running tasks.
- Transport Support: Supports multiple transport types including stdio, streamable-http, and SSE.
- Configuration: Easily configurable via environment variables for different transport types and settings.
Tools
register_agent
Register an A2A agent with the bridge server.
list_agents
Get a list of all registered agents.
unregister_agent
Remove an A2A agent from the bridge server.
send_message
Send a message to an agent and get a task_id for the response.
get_task_result
Retrieve a task's result using its ID.