Compose router, evaluator, and handoff workflows over MCP servers with mcp-agent

Connect LLMs to MCP servers through composable patterns like router, evaluator-optimizer, and orchestrator flows without hand-managing server lifecycles.

Prerequisites

Python runtime, MCP servers, LLM API key, optional Temporal for durable workflows

Installation

Use the upstream install or setup path that matches your environment:

• uv init
• uv add "mcp-agent[openai]"
• uv run main.py
• uv add "mcp-agent"

Requirements and caveats from upstream:

• python
• We recommend using uv to manage your Python projects (uv init).

Basic usage or getting-started notes:

• <a id="minimal-example"></a>

• Minimal example

• async with app.run():

• Source: https://github.com/lastmile-ai/mcp-agent

• Extracted from upstream docs: https://raw.githubusercontent.com/lastmile-ai/mcp-agent/HEAD/README.md

Documentation

• https://docs.mcp-agent.com/

Source

• Agent Skill Exchange