interactive-feedback-mcp

bulice/interactive-feedback-mcp

3.3

If you are the rightful owner of interactive-feedback-mcp and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

Interactive Feedback MCP is a server designed for AI-assisted development tools, enabling real-time human-machine interaction and feedback collection.

Interactive Feedback MCP is a server based on the Model Context Protocol (MCP), specifically designed to enhance AI-assisted development tools like Cursor, Cline, and Windsurf. This project builds upon the original design by Fábio Ferreira and incorporates modern implementations from the mcp-feedback-collector project. It provides a graphical interface that allows users to offer textual feedback, upload images, execute commands, and view outputs in real-time, fostering a collaborative development environment. The server is built using Python 3.11+, FastMCP, PySide6, psutil, and Pillow, ensuring a robust and flexible architecture. The project structure is well-organized, with separate components for the MCP server, GUI interface, diagnostic tools, and testing scripts. The server supports both dark and light themes, responsive design, and project-specific settings, making it adaptable to various user preferences and project requirements.

Features

  • Interactive Feedback: Provides detailed text feedback, supports image uploads, and enables real-time interaction with AI assistants.
  • User Interface: Offers dark and light themes, dynamic theme switching, and responsive design for different environments.
  • Command Execution: Displays real-time command execution results, monitors process status, and saves command history.
  • Project Management: Supports project-specific settings, configuration persistence, and flexible startup options.
  • Technical Stack: Utilizes Python 3.11+, FastMCP, PySide6, psutil, and Pillow for a robust and flexible architecture.

Usages

usage with Cursor

{
  "mcpServers": {
    "interactive-feedback-mcp": {
      "command": "/Users/lee/tmp/interactive-feedback-mcp/mcp_server.sh",
      "args": [],
      "timeout": 600,
      "autoApprove": [
        "interactive_feedback"
      ]
    }
  }
}

Tools

  1. interactive_feedback

    Tool for providing interactive feedback to AI assistants.

  2. get_image_info

    Tool for retrieving information about uploaded images.