The Agent Skills Directory

[SAFE]: The skill consists of markdown-based instructions for product acceptance testing methodology. It does not include any scripts, executable files, or tool configurations that could perform unauthorized actions.
[INDIRECT_PROMPT_INJECTION]: The skill defines a process for analyzing external data (proof records and test evidence), which creates an inherent attack surface for indirect prompt injection. However, as the skill lacks automated execution capabilities or access to sensitive tools, this represents a structural characteristic of reasoning tasks rather than a malicious vulnerability.

product-acceptance