mcp-agent
Build effective agents using Model Context Protocol and simple workflow patterns
- 8.6k stars
- 893 forks
- 149 open issues
- Language: Python
- License: Apache-2.0
- updated
Install
local · uvx (python)claude mcp add mcp-agent -- uvx mcp-agent{
"mcpServers": {
"mcp-agent": {
"command": "uvx",
"args": ["mcp-agent"]
}
}
}{
"mcpServers": {
"mcp-agent": {
"command": "uvx",
"args": ["mcp-agent"]
}
}
}- Run the command above in your terminal (Claude Code), or paste the JSON config into claude_desktop_config.json (Claude Desktop).
- Replace any <placeholder> values with your API keys or paths.
- Restart Claude. The MCP server and its tools appear automatically.
Package name inferred from the repository name. Verify it exists on PyPI, or clone https://github.com/lastmile-ai/mcp-agent and follow its README.
Editor's note
mcp-agent is a Python framework for building LLM-driven agents entirely on top of the Model Context Protocol, handling the full MCP lifecycle including tools, resources, prompts, notifications, OAuth, sampling, elicitation, and roots. It implements every workflow pattern described in Anthropic's "Building Effective Agents" guide, including orchestrator, evaluator-optimizer, router, and map-reduce, and these patterns can be composed together into more complex pipelines. Agents connect to any MCP server by name and can be exposed as MCP servers themselves using a FastMCP-compatible API, making them consumable by Claude Desktop or other MCP clients. The framework works with multiple LLM backends including the Anthropic API and OpenAI. A notable feature is optional Temporal integration for durable execution, which allows long-running agent workflows to pause, resume, and recover from failures without any changes to the agent API. Python developers building multi-step AI workflows, teams deploying production agents, and anyone who wants to treat MCP as the sole interface layer for agent construction are the primary audience.
MCP Servers overview
What people ask about mcp-agent
What is lastmile-ai/mcp-agent?
lastmile-ai/mcp-agent is mcp servers for the Claude AI ecosystem. Build effective agents using Model Context Protocol and simple workflow patterns It has 8.6k GitHub stars and its last recorded update is dated 2026-01-25.
How do I install mcp-agent?
You can install mcp-agent by cloning the repository (https://github.com/lastmile-ai/mcp-agent) or following the README instructions on GitHub. ClaudeWave also provides quick install blocks on this page.
Is lastmile-ai/mcp-agent safe to use?
Our security agent has analyzed lastmile-ai/mcp-agent and assigned a Trust Score of 82/100 (tier: Trusted). See the full breakdown of passed checks and flags on this page.
Who maintains lastmile-ai/mcp-agent?
lastmile-ai/mcp-agent is maintained by lastmile-ai. The last recorded GitHub activity is dated 2026-01-25, with 149 open issues.
Are there alternatives to mcp-agent?
Yes. On ClaudeWave you can browse similar mcp servers at /categories/mcp, sorted by popularity or recent activity.