The Agent Skills Directory

[SAFE]: The skill instructions and protocols are focused on defensive security auditing and quality standards. No malicious intent, obfuscation, or unauthorized data access patterns were identified.\n- [COMMAND_EXECUTION]: The skill utilizes well-known security tools, including npm audit, bandit, and lighthouse. These commands are used as intended for auditing source code and assessing performance, posing no security threat to the agent's environment.\n- [PROMPT_INJECTION]: The core rules and execution protocols are strictly related to QA tasks and do not attempt to override safety guidelines or bypass established AI behavior constraints.\n- [SAFE]: While the skill reviews external code (presenting an indirect prompt injection surface), it manages this risk through structured analysis protocols and the use of specialized auditing tools, which is consistent with its primary purpose.

qa-agent