The Agent Skills Directory

[SAFE]: The skill provides instructions for a code-review persona focused on failure modes, error handling, and timeouts. It does not contain any malicious code, command execution patterns, or data exfiltration attempts.
[SAFE]: No hardcoded credentials or sensitive file paths were found. The skill does not initiate unauthorized network connections.
[SAFE]: The skill metadata and instructions are consistent with its stated purpose, and there is no evidence of prompt injection or obfuscation techniques.

reliability-reviewer