ads-generate
Pass
Audited by Gen Agent Trust Hub on Aug 20, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data (creative briefs and source images) which constitutes an attack surface. However, it provides explicit defensive instructions to mitigate this risk.
- Ingestion points: Creative briefs and brand profiles mentioned in Step 1 of SKILL.md.
- Boundary markers: Step 2 explicitly instructs the agent to treat inputs as untrusted and ignore embedded instructions.
- Capability inventory: The skill manages image generation via external providers and writes assets to local directories.
- Sanitization: Step 4 and 5 include validation steps for policies, redacted prompt summaries, and safe-zone consistency.
- [DATA_EXPOSURE]: The skill follows security best practices for data handling. It specifically mandates that provider credentials should not be exposed and that local filesystem paths must be redacted in favor of relative locators.
Audit Metadata