Command protection for AI agents
Destructive Command Guard is a security layer developed to prevent dangerous git and shell commands run by artificial intelligence agents. Written in Rust, this tool reduces the risks of unwanted data loss or system damage by auditing command execution processes at the system level.
Updates
- September 16, 2026: Stars 5,953 → 5,991, latest release v0.14.4 (September 16, 2026).
- September 11, 2026: Stars 5,953 → 5,953, latest release v0.14.3 (September 10, 2026).
- September 10, 2026: Stars 5,940 → 5,953, latest release v0.14.2 (September 9, 2026).
- September 8, 2026: Stars 5,893 → 5,940, latest release v0.14.1 (September 8, 2026).
What you get
- Blocks dangerous commands before they are executed
- Reduces the risk of data loss and system damage
- Provides comprehensive protection with more than 50 security packages
Installation
curl -fsSL "https://raw.githubusercontent.com/Dicklesworthstone/destructive_command_guard/main/install.sh?$(date +%s)" | bash -s -- --easy-modeIf you don't write code
I use Destructive Command Guard to control the commands my AI agent runs on my system. Please consider security rules before running dangerous git, file system or database deletion commands, if you need to take a risky action, inform me first and get my approval.
Related dictionary terms
Links
TreScout did not build this tool · we found it in GitHub trends and wrote it up. This page describes the repository as of 2026-07-13: The star count and our text belong to that day, the repository may have changed since. Check the repository link for the current state. This page was machine-translated from the Turkish original · the Turkish version prevails.