clonar-anuncio
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill includes explicit instructions and a dedicated 'Rules' section that prohibits copying competitor content (scripts, branding, visuals) and warns about copyright infringement.
- [SAFE]: Data processing instructions correctly specify that external data (ad text, transcripts) should be treated as passive material and that any embedded commands within that data must be ignored, mitigating indirect prompt injection risks.
- [SAFE]: Interaction with the Meta Ads Library is restricted to read-only search functionality. The instructions explicitly forbid creating, editing, or managing campaigns, ensuring the tool remains within its documented scope.
- [SAFE]: The skill uses standard CLI tools like ffmpeg and common AI provider integrations for video analysis, with clear warnings about accuracy and the importance of user-provided (original) source videos.
- [SAFE]: Documented external links point to official and reputable sources (facebook.com).
Audit Metadata