This project provides an Interactive Feedback MCP server designed to facilitate a human-in-the-loop workflow for AI-assisted development tools like Cursor. It enables users to run commands, view their output, and provide textual feedback directly to the AI, reducing unnecessary tool calls and saving resources.

How It Works

The server acts as an intermediary, allowing AI assistants to pause execution and solicit user feedback via a dedicated interactive_feedback tool call. This approach guides the AI to check in with the user rather than making speculative, high-cost tool invocations, consolidating multiple potential AI interactions into a single, feedback-driven request.

Quick Start & Requirements

• Install: Clone the repository, navigate to the directory, and run uv sync to install dependencies and create a virtual environment.
• Run: Execute uv run server.py.
• Prerequisites: Python 3.11+, uv package manager.
• Configuration: Requires manual configuration within AI tools (e.g., Cursor's MCP settings) or via a mcp.json file, pointing to the server's execution path.
• Docs: dotcursorrules.com

Highlighted Details

• Reduces premium AI request costs by consolidating feedback loops.
• Compatible with Cursor, Cline, and Windsurf.
• Stores project-specific configurations using Qt's QSettings.
• Offers a development mode with a web interface for testing (uv run fastmcp dev server.py).

Maintenance & Community

• Developed by Fábio Ferreira (@fabiomlferreira).
• Contact and support available via X (@fabiomlferreira).

Licensing & Compatibility

• The license is not explicitly stated in the README.

Limitations & Caveats

• The README does not specify the project's license, which may impact commercial use or closed-source integration.