Interactive Feedback

Enables human-in-the-loop workflows by launching a browser interface where users can provide interactive feedback during task exec

zivhdinfo 28
Claude CodeClaude DesktopGeneric
View source ↗

Interactive Feedback MCP provides a Node.js-based server that enables AI assistants to request human feedback during task execution through a modern web interface. The implementation launches a browser-based UI where users can run commands, view console output in real-time via WebSocket connections, and provide interactive feedback that gets returned to the AI assistant as structured JSON data. Built with Express, WebSocket support, and cross-platform process management, it's designed for human-in-the-loop workflows where AI assistants need user validation, approval, or additional input before proceeding with tasks, making it valuable for code review processes, deployment confirmations, and iterative development scenarios.

Source

Repository: https://github.com/zivhdinfo/interactive-feedback-mcp-nodejs

Maintain Interactive Feedback?

Let people know it's listed here — add the badge (live metrics, light/dark aware) or a plain link to your README or docs.

[Interactive Feedback on getagentictools](https://getagentictools.com/mcp/zivhdinfo-interactive-feedback-mcp-nodejs?ref=badge)