loading…
Search for a command to run...
loading…
Enables AI assistants to request user feedback and confirmation to prevent speculative tool calls and reduce resource usage. It streamlines task completion by c
Enables AI assistants to request user feedback and confirmation to prevent speculative tool calls and reduce resource usage. It streamlines task completion by consolidating multiple potential steps into a single, feedback-aware process.
By guiding the AI assistant to check in with the user instead of branching out into speculative, high-cost tool calls, this module can drastically reduce the number of premium requests. In some cases, it helps consolidate what would be up to 25 tool calls into a single, feedback-aware request — saving resources and improving performance.
One command to clone, install, and configure everything automatically for Antigravity IDE:
Windows (PowerShell):
irm https://raw.githubusercontent.com/nhatpse/Antigravity-MCP/v1.3.3/install.ps1 | iex
Linux / macOS:
curl -fsSL https://raw.githubusercontent.com/nhatpse/Antigravity-MCP/v1.3.3/install.sh | bash
The installer will: clone the repo → install dependencies → configure MCP server → add coding rules. Just restart Antigravity after it finishes.
uv will download and manage the required Python version automatically!irm https://astral.sh/uv/install.ps1 | iexcurl -LsSf https://astral.sh/uv/install.sh | shClone or download this repository.
Install dependencies:
cd path/to/interactive-feedback-mcp
uv sync
Add the MCP server to your Antigravity configuration (~/.gemini/antigravity/mcp.json):
{
"mcpServers": {
"interactive-feedback-mcp": {
"command": "uv",
"args": [
"--directory",
"/path/to/interactive-feedback-mcp",
"run",
"server.py"
]
}
}
}
Note: If
uvis not in your system PATH, use the full path to theuvexecutable instead (e.g.,C:\\Users\\<user>\\AppData\\Local\\Python\\...\\Scripts\\uv.exe).
For the best results, add the following as a coding rule in your AI assistant:
Whenever you want to ask a question, always call the MCP
interactive_feedback.
Whenever you're about to complete a user request, call the MCPinteractive_feedbackinstead of simply ending the process. Keep calling MCP until the user's feedback is empty, then end the request.

To run the server in development mode with a web interface for testing:
uv run fastmcp dev server.py
Created by nhatpse.
This project is licensed under the MIT License.
Добавь это в claude_desktop_config.json и перезапусти Claude Desktop.
{
"mcpServers": {
"interactive-feedback-mcp": {
"command": "npx",
"args": []
}
}
}