Prefect
Interact with the Prefect API for workflow orchestration and management.
Prefect MCP Server
This repository provides a Prefect MCP server configuration using the prefect-mcp-server package with a reliable running mechanism via uvx. The configuration is tailored for use with the Cursor IDE.
Prerequisites
- Python 3.9 or newer.
- A preferred virtual environment tool (such as uv) for managing Python environments.
- Prefect 3 (see Prefect Documentation for installation instructions).
Installation
Create and activate your virtual environment, then install Prefect MCP Server:
uv venv --python 3.12 && source .venv/bin/activate
uv pip install -U prefect-mcp-server
Configuration
The server is configured via the .cursor/mcp.json file. The updated configuration is as follows:
{
"mcpServers": {
"prefect": {
"command": "uvx",
"args": [
"prefect-mcp-server"
],
"env": {}
}
}
}
This configuration ensures that the server uses the uvx command with the exact package version installed via uv pip install. This approach provides enhanced reliability and consistency in your development environment.
Environment Variables
Set the following environment variables to configure your Prefect environment. You can create a file named .env in the project root with entries such as:
PREFECT_API_URL=http://localhost:4200/api
Additionally, if needed, set other environment variables like PREFECT_API_KEY to authenticate with your Prefect server or Prefect Cloud.
Running the Server
To start the server, you can run the following command:
uv run <script>
Alternatively, if you are using the Cursor IDE with its configuration, the server will be automatically invoked with the command specified in .cursor/mcp.json.
Documentation
Detailed documentation on the Prefect MCP Server functionality and usage is available in the docs/prefect_mcp_documentation.md file. The documentation includes:
- Complete list of available tools and their parameters
- Instructions for installation and configuration
- Examples of usage with different MCP clients
- Prefect 3.0 compatibility information
Cursor Rules
This repository includes Cursor Rules for working with the Prefect MCP Server, located in the .cursor/rules/ directory. These rules provide contextual help and guidance when working with Prefect MCP in the Cursor IDE.
Additional Information
- For further details on Prefect installation and usage, please refer to the Prefect Documentation.
- For information about the Model Context Protocol (MCP), see the MCP Documentation.
- Use
uv runfor running scripts within the configured environment as recommended by Cursor.
Happy coding!
Serveurs connexes
Scout Monitoring MCP
sponsorPut performance and error data directly in the hands of your AI assistant.
Alpha Vantage MCP Server
sponsorAccess financial market data: realtime & historical stock, ETF, options, forex, crypto, commodities, fundamentals, technical indicators, & more
Muster
A universal control plane for managing MCP servers and providing intelligent tool discovery for AI agents.
Revit MCP
Interact with Autodesk Revit using the MCP protocol. This server provides AI tools and requires the corresponding Revit plugin to function.
Remote MCP Server on Cloudflare
A remote MCP server deployable on Cloudflare Workers with OAuth login support and local development capabilities.
Multiverse MCP Server
A middleware server for running multiple, isolated instances of MCP servers with unique namespaces and configurations.
OpenAPI2MCP
Converts OpenAPI specifications into MCP tools, enabling AI clients to interact with external APIs seamlessly.
Windows CLI
Interact with Windows command-line interfaces like PowerShell, CMD, Git Bash, and WSL.
Graph Tools
An interactive graph analysis toolkit with web visualizations and AI-powered analysis capabilities.
zig-mcp
MCP server for Zig that connects AI coding assistants to ZLS (Zig Language Server) via LSP — 16 tools for code intelligence, build, and test.
PCM
A server for reverse engineering tasks using the pcm toolkit. Requires a local clone of the pcm repository.
Copado MCP
A remote MCP server deployable on Cloudflare Workers without authentication.
