Feedback Collector

v1.0.0Communicationstable

Enables AI assistants to collect interactive user feedback through a modern GUI that supports text input, multiple image uploads (via file selection or clipboard paste), allowing users to provide comprehensive feedback with screenshots and comments.

mcp-feedback-collectormcpai-integration
Share:
239
Stars
0
Downloads
0
Weekly
0/5

What is Feedback Collector?

Feedback Collector is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to enables ai assistants to collect interactive user feedback through a modern gui that supports text input, multiple image uploads (via file selection or clipboard paste), allowing users to provide comp...

Enables AI assistants to collect interactive user feedback through a modern GUI that supports text input, multiple image uploads (via file selection or clipboard paste), allowing users to provide comprehensive feedback with screenshots and comments.

This server falls under the Communication category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Enables AI assistants to collect interactive user feedback t

Use Cases

Collect interactive user feedback through a modern GUI.
Support text input and multiple image uploads from users.
Capture screenshots and comments for comprehensive feedback.
zhaobu

Maintainer

LicenseMIT License
Languagepython
Versionv1.0.0
UpdatedMay 21, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx mcp-feedback-collector

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Feedback Collector

MCP Feedback Collector is a Python-based MCP server that allows AI assistants to pause their workflow and solicit rich, structured feedback from users through a native GUI dialog. Instead of ending a task and waiting for the user's next message, the AI calls the collect_feedback tool which opens a 700×800 px always-on-top window where users can type comments and attach multiple images via file selection or clipboard paste. This is especially useful for iterative design, code review, or any agentic workflow where the AI needs human confirmation or screenshots before proceeding.

Prerequisites

  • Python 3.10 or later installed
  • uv package manager installed (pip install uv or curl -LsSf https://astral.sh/uv/install.sh | sh)
  • An MCP-compatible AI client such as Claude Desktop
  • A desktop environment (GUI is required — headless servers are not supported)
1

Install uv if not already present

MCP Feedback Collector is installed and run via uvx, which is part of the uv package manager. Install uv first if you have not already done so.

pip install uv
# or on macOS/Linux:
curl -LsSf https://astral.sh/uv/install.sh | sh
2

Verify the package runs correctly

Test that uvx can fetch and launch the mcp-feedback-collector package. If it starts without errors the installation path is working. Press Ctrl+C to stop it.

uvx mcp-feedback-collector
3

Open your Claude Desktop configuration file

Locate and open claude_desktop_config.json. On macOS the path is ~/Library/Application Support/Claude/claude_desktop_config.json. On Windows it is %APPDATA%\Claude\claude_desktop_config.json.

4

Add the server entry with the recommended timeout

Add the mcp-feedback-collector server to the mcpServers block. Set MCP_DIALOG_TIMEOUT to 600 seconds so users have enough time to compose detailed feedback with screenshots. The default of 300 seconds is often too short for complex feedback.

{
  "mcpServers": {
    "mcp-feedback-collector": {
      "command": "uvx",
      "args": ["mcp-feedback-collector"],
      "env": {
        "PYTHONIOENCODING": "utf-8",
        "MCP_DIALOG_TIMEOUT": "600"
      }
    }
  }
}
5

Restart Claude Desktop

Save the configuration file and fully restart Claude Desktop. The server will appear under Available MCP Servers in Settings once connected successfully.

6

Configure your AI assistant to collect feedback

Instruct the AI assistant to call collect_feedback instead of ending each task. A recommended system prompt pattern ensures the assistant keeps requesting feedback until the user submits an empty response.

Feedback Collector Examples

Client configuration

Complete Claude Desktop configuration for MCP Feedback Collector with a 600-second dialog timeout and UTF-8 encoding for international text.

{
  "mcpServers": {
    "mcp-feedback-collector": {
      "command": "uvx",
      "args": ["mcp-feedback-collector"],
      "env": {
        "PYTHONIOENCODING": "utf-8",
        "MCP_DIALOG_TIMEOUT": "600"
      }
    }
  }
}

Prompts to try

Example prompts that demonstrate how to use the feedback collector during agentic workflows.

- "After you finish generating the UI mockup, use the feedback collector to ask me for my thoughts before continuing"
- "Build the first draft of the report, then collect my feedback — keep iterating until I submit an empty response"
- "Redesign this landing page, then pause and let me paste a screenshot showing which section needs changing"
- "Complete the code refactor, then ask me via the feedback GUI if there are any edge cases you missed"

Troubleshooting Feedback Collector

The feedback dialog does not appear when collect_feedback is called

Ensure you are running in a desktop environment with a display. Check that DISPLAY is set on Linux (e.g., export DISPLAY=:0). The GUI uses tkinter which requires a graphical session — SSH without X forwarding will not work.

Dialog times out before the user can finish writing feedback

Increase MCP_DIALOG_TIMEOUT in the env block of your config. The default is 300 seconds; set it to 600 or higher. Restart Claude Desktop after changing the value.

Pasted images do not appear in the feedback dialog

Use Ctrl+V (or Cmd+V on macOS) inside the dialog to paste clipboard images. Supported formats are PNG, JPG, JPEG, GIF, BMP, and WebP. If paste does not work, use the file selection button instead to browse for your screenshot.

Frequently Asked Questions about Feedback Collector

What is Feedback Collector?

Feedback Collector is a Model Context Protocol (MCP) server that enables ai assistants to collect interactive user feedback through a modern gui that supports text input, multiple image uploads (via file selection or clipboard paste), allowing users to provide comprehensive feedback with screenshots and comments. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Feedback Collector?

Follow the installation instructions on the Feedback Collector GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Feedback Collector?

Feedback Collector works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Feedback Collector free to use?

Yes, Feedback Collector is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.

Browse More Communication MCP Servers

Explore all communication servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "mcp-feedback-collector": { "command": "npx", "args": ["-y", "mcp-feedback-collector"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Feedback Collector?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides