claude-code-hooks
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides templates and documentation for creating Claude Code hooks, which are a legitimate feature of the Claude Code environment.
- [SAFE]: The provided scripts use standard system utilities (bash, python3, git, grep) to process tool-call events.
- [SAFE]: The skill emphasizes security best practices, such as using shlex for shell tokenization to prevent command injection bypasses and using human-in-the-loop gates (via osascript dialogs) for irreversible actions.
- [SAFE]: No evidence of data exfiltration, credential harvesting, or malicious obfuscation was found. The use of Python embedded in Bash scripts is for JSON parsing, which is a standard requirement for the hook protocol.
Audit Metadata