The Agent Skills Directory

[SAFE]: The skill instructions do not contain any malicious patterns, obfuscation, or unauthorized access attempts.
[DATA_EXPOSURE]: The skill reads from local reference files (ads/references/platform-specs.md, ads/references/benchmarks.md, and ads/references/scoring-system.md), which is consistent with its stated purpose of auditing creative quality against established benchmarks.
[INDIRECT_PROMPT_INJECTION]: The skill processes external ad performance data and creative assets. While this represents an ingestion point for untrusted data, the skill lacks any dangerous capabilities (such as network access, file writing, or command execution) that could be exploited via indirect injection, resulting in a safe profile.

ads-creative