The Agent Skills Directory

[SAFE]: The skill defines a senior support QA persona and provides a checklist for reviewing customer responses. It evaluates user-provided drafts for accuracy, completeness, tone, and clarity based on a provided checklist.\n- [SAFE]: The skill uses the $ARGUMENTS placeholder to process user-provided text. While the input is interpolated directly into the prompt without explicit delimiters, the skill lacks access to sensitive data or dangerous tools, and model invocation is explicitly disabled via the 'disable-model-invocation: true' setting, which neutralizes risks from prompt injection.\n- [SAFE]: No suspicious patterns such as obfuscation, data exfiltration, or remote code execution were detected. The skill is entirely declarative and focuses on providing feedback on text.

qa-response