product-ad-video-script
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions and associated framework are focused on generating marketing scripts from provided product data. No malicious patterns, data exfiltration attempts, or unauthorized command executions were found. The skill emphasizes safety and factual accuracy in the generated advertisements.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from external sources such as landing pages and product descriptions. It includes a structured 'Fact Sheet' methodology and a 'Claim Safety Gate' to verify information before output, reducing the risk of unintended behavior from processing untrusted content. The capability is limited to content planning and scripting.
- [REMOTE_CODE_EXECUTION]: No remote code execution patterns, unauthorized package installations, or piped shell commands were detected. The skill uses local reference files to guide its behavior.
Audit Metadata