mira-product
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data provided by the user to generate media prompts.
- Ingestion points: The skill takes product descriptions and exact label copy from the user (SKILL.md).
- Boundary markers: The instructions specify that the user's own label copy must be reproduced "word for word," which increases the risk of the agent obeying instructions hidden within the label text during prompt construction.
- Capability inventory: The skill uses
generate_imageandgenerate_videotools to produce output based on the user-provided content. - Sanitization: There are no explicit sanitization steps or delimiters mentioned for the user-supplied text before it is interpolated into the final generation instructions.
Audit Metadata