mcp-agent

Name: mcp-agent
Availability: InStock
Author: lastmile-ai

by lastmile-ai

Composable framework to build effective agents using Model Context Protocol with durable execution and advanced patterns.

0 stars

0 views

Works in:Cursor

Exposes:ToolsResourcesPrompts

What it does\n`mcp-agent` is a Python framework designed to build robust AI agents by leveraging the Model Context Protocol (MCP). It allows developers to connect LLMs to various MCP servers and implement complex agentic patterns such as map-reduce, router, and evaluator-optimizer without writing boilerplate code.\n\n## Tools\n- Agent Orchestration: Coordinate multiple worker agents and manage their lifecycles.\n- Workflow Patterns: Implement parallel processing, intent classification, and deep research pipelines.\n- Durable Execution: Integration with Temporal for pause, resume, and recovery of agent states.\n\n## Installation\nAdd the following to your `claude_desktop_config.json`:\n\n`json\n{\n "mcpServers": {\n "mcp-agent": {\n "command": "uvx",\n "args": ["mcp-agent"]\n }\n }\n}\n`\n\n## Supported hosts\n- Claude Desktop\n- Cursor\n- VS Code\n

uvx mcp-agent

Choose your AI client and follow the steps below.

Cursor

Add to mcp.json in ~/.cursor/ or project .cursor/

Claude Desktop

{\n  "mcpServers": {\n    "mcp-agent": {\n      "command": "uvx",\n      "args": ["mcp-agent"]\n    }\n  }\n}