The Agent Skills Directory

[SAFE]: The skill defines a comprehensive framework for auditing the safety and correctness of hook scripts. It specifically looks for and flags dangerous patterns like REMOTE_CODE_EXECUTION (e.g., curl | sh) and COMMAND_EXECUTION (e.g., rm -rf) in analyzed files.
[SAFE]: No instances of DATA_EXFILTRATION, CREDENTIALS_UNSAFE, or unauthorized network activity were detected. The skill interacts with local project files for analysis purposes only.
[SAFE]: There is no evidence of obfuscation or hidden payloads. The logic for argument parsing and workflow execution is transparent and follows standard practices.
[SAFE]: The skill facilitates security best practices by recommending safety headers (e.g., set -euo pipefail) and proper exit codes for blocking malicious actions.

enhance-hooks