Armory
Source
Browse
Hooks

ai-bash-guard

AI-powered bash command security guard. Before any Bash command runs, a lightweight Claude subagent evaluates it for destructive or irreversible patterns — recursive deletes, force pushes to protected branches, database drops, and credential exposure — and blocks execution with a clear explanation if flagged. Uses PreToolUse with type:agent, which is the only hook pattern that can block tool execution via AI reasoning.

Score
Unranked
Evidence
No signals yet
Last commit
Not known
Listed

Install

armory install ai-bash-guard --cli claude

writes the file to.claude/hooks/ai-bash-guardListed as compatible

Configuration
# fetches the hook script into:
.claude/hooks/ai-bash-guard
# then register it in your settings.json "hooks" block.

↳ Add the hook entry to settings.json, then restart the harness.

Needs the armory CLI · not on npm yet, build it from cli/ in the repository

What it is

AI-powered bash command security guard. Before any Bash command runs, a lightweight Claude subagent evaluates it for destructive or irreversible patterns — recursive deletes, force pushes to protected branches, database drops, and credential exposure — and blocks execution with a clear explanation if flagged. Uses PreToolUse with type:agent, which is the only hook pattern that can block tool execution via AI reasoning.

When to use it

AI-powered bash command security guard. Before any Bash command runs, a lightweight Claude subagent evaluates it for destructive or irreversible patterns — recursive deletes, force pushes to protected branches, database drops, and credential exposure — and blocks execution with a clear explanation if flagged. Uses PreToolUse with type:agent, which is the only hook pattern that can block tool execution via AI reasoning.

How to install / invoke

# Wire this hook script into .claude/settings.json
curl -sL https://raw.githubusercontent.com/davila7/claude-code-templates/main/cli-tool/components/hooks/security/ai-bash-guard.json -o .claude/hooks/ai-bash-guard.json

Notes

Extracted from davila7/claude-code-templates, security category.