Product Hunt! 👋 I'm prit , and I built MCP Guard because connecting AI agents to databases terrified me. The Model Context Protocol is powerful, but one hallucinated query can delete your entire database. I've seen it happen to others, and I didn't want it to happen to me. MCP Guard is a hosted security dashboard that sits between your AI agents and your MCP servers. It blocks dangerous commands automatically, logs every request, and gives you visibility into what your AI is actually doing. What it does: Blocks DROP, DELETE without WHERE, rm -rf automatically Shows you every request your AI agents make Lets you configure permissions per agent No CLI or local proxy—just a browser dashboard Why this exists: I wanted to use AI agents for development without worrying about accidental data destruction. I couldn't find a simple solution, so I built one. Check it out: https://mcp-shield.vercel.app Questions? Feedback? Let me know! I'm here all day to answer. 🙏

MCP Guard

The firewall for AI agents. Block dangerous commands

The firewall for AI agents. Block dangerous commands