creativeclaw-create-ugc-ad
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user content such as product briefs and claims which influence the generated scripts and media.
- Ingestion points: User briefs and claims are processed in SKILL.md.
- Boundary markers: No explicit structural delimiters are defined for user input.
- Capability inventory: The skill utilizes project management, video generation, and audio assembly tools.
- Sanitization: The instructions require the agent to remove or qualify unverified performance, health, or financial claims.
Audit Metadata