strands-agents-mcp

imgaray/strands-agents-mcp

3.3

If you are the rightful owner of strands-agents-mcp 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.

A Model Context Protocol (MCP) server for executing Strands agents, providing integration with Amazon Q and other MCP-compatible systems.

Strands Agent MCP is a bridge between the Strands agent framework and the Model Context Protocol (MCP). It allows users to register Strands agents as MCP tools, execute them through MCP, and find agents by specific skills. The project employs a plugin architecture, enabling easy addition of new agents without modifying the core code. This server is currently in the alpha stage and not yet available on PyPI, requiring installation from source. The server supports environment variables for custom plugin paths and namespaces, and it can be connected to Amazon Q for enhanced functionality. The architecture consists of a server, registry, and plugins, with automatic discovery and registration of agents.

Features

  • Register Strands agents as MCP tools
  • Execute Strands agents through MCP
  • Find agents by specific skills
  • Plugin architecture for easy agent addition
  • Integration with Amazon Q

Usages

usage with local development

bash
strands-agent-mcp

usage with amazon q

python
# Connect to Amazon Q using the MCP server
# Refer to Amazon Q documentation for connection parameters
# Example usage:
# execute_agent(agent_name='my-agent', prompt='Hello, world!')
# list_agents()

Tools

  1. execute_agent

    Execute an agent with parameters agent_name and prompt

  2. list_agents

    List all available agents