git-guardrails-claude-code
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a security tool designed to protect the user's environment. It operates entirely through local scripts and configuration changes to Claude Code's hook system.
- [INDIRECT_PROMPT_INJECTION]: The hook logic uses structured parsing to identify and block destructive commands, serving as a mitigation against agents accidentally performing high-risk git operations. The implementation includes measures to prevent bypasses through shell comments or heredocs.
Audit Metadata