7 layers between your AI agents and disaster.

Guardclaw - 7 layers between your AI agents and disaster.

by•16d ago

Hello Everyone! Hope you’re well! We just opened GuardClaw public beta. It sits between AI agents and tool access (terminal, files, MCP servers) and blocks risky actions before they run. It is local-first, rule-based, and has 0 LLMs in the security path. Stats: 7 defense layers, 1,500+ detection patterns across 11 attack categories. Works with Claude Code, Codex CLI, OpenClaw, Claude Desktop, plus Docker/MCP setups. Try it and tell us what breaks or feels missing!

Replies

Best

Maker

📌

You can succeed faster if you can work without worrying about things going wrong. One of the biggest challenges we’ve had is trying to figure out how to stay safe when using AI agents (especially personal/ offline agents) and setting up access. There weren’t many tools that could stop agents actions and work off-line, so we built our first app and we would love any and all feedback! Thank you for trying it! We hope this can help you as much as it is helping us

Report

16d ago