The Agent Skills Directory

[COMMAND_EXECUTION] (SAFE): The skill references running commands like npm audit, bandit, and lighthouse for security and performance auditing. These are standard industry tools for QA and security analysis.
[PROMPT_INJECTION] (SAFE): There are no attempts to override agent behavior or bypass safety filters. The instructions focus on improving the accuracy of security audits and preventing false positives.
[DATA_EXFILTRATION] (SAFE): The skill does not access sensitive local paths or perform network requests to untrusted domains. It focuses on local code analysis.
[EXTERNAL_DOWNLOADS] (SAFE): No external downloads or remote script executions are present.
[REMOTE_CODE_EXECUTION] (SAFE): No patterns for dynamic code execution (eval, exec) using external or untrusted data were found.

qa-agent