adapting-formats
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes user-supplied images and URLs, creating an attack surface where malicious instructions embedded in the source data could influence the agent's content generation process.
- Ingestion points: User-provided ad images and URLs (SKILL.md, Step 1).
- Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands within the analyzed content.
- Capability inventory: The skill uses
image_analysisto read content andimage_generateto create new images based on that content (SKILL.md, Steps 2 and 7). - Sanitization: There is no evidence of sanitization or validation of the text extracted from the source images before it is used in the prompt construction.
Audit Metadata