Prerequisites
Python 3.10+
mcp-agent targets Python 3.10 or newer.
Node.js (optional)
Some MCP servers (e.g. filesystem) are distributed via npx and require Node.js.
Run the CLI with uvx
You can run mcp-agent commands without a global install by using uvx:
uvx mcp-agent deploy …, you can paste it directly into your terminal. If you prefer a persistent install, pip install mcp-agent still works—but uvx keeps things simple.
Add mcp-agent to your project
Inside your project directory, initialize a new project and add mcp-agent as a dependency:uv add writes to pyproject.toml and uv.lock, which the project templates (uvx mcp-agent init …) expect.
Provider extras
mcp-agent includes optional extras for different LLM providers. Install the ones you need:OpenAI
OpenAI
Anthropic
Anthropic
Azure OpenAI
Azure OpenAI
AWS Bedrock
AWS Bedrock
Google Gemini
Google Gemini
All Providers
All Providers
Verify your setup
1
Create a project folder
2
Install dependencies
3
Smoke test
Refer to the CLI reference for the complete command list, then head to the Quickstart to scaffold a project with
uvx mcp-agent init.Next steps
- Quickstart – create a finder agent and run it locally.
- Deploy to Cloud – publish your agent with
uvx mcp-agent deploy. - MCP Servers – learn how FastMCP and other servers plug into your agents.
