git-guardrails-claude-code

Pass

Audited by Gen Agent Trust Hub on Jul 7, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses shell scripts and configuration modifications to implement a security feature.
  • The script scripts/block-dangerous-git.sh is designed to be installed as a PreToolUse hook in Claude Code settings (.claude/settings.json).
  • It requires the user to run chmod +x to make the script executable, which is standard for local tool installation.
  • The script uses jq to parse tool input and grep to validate commands against a predefined list of destructive Git patterns (e.g., git push, git reset --hard).
  • There is no evidence of network exfiltration, credential harvesting, or obfuscation. All operations are local to the user's repository or home directory.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 7, 2026, 06:17 PM
Security Audit — agent-trust-hub — git-guardrails-claude-code