product-studio-visual-by-tp9kyk
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes content from an external source file to define the output prompt's style and structure.
- Ingestion points: The agent is instructed to read
references/cases.mdon every invocation to select an anchor case. - Boundary markers: A mandatory "reference contract" using "Preserve / Replace / Avoid" tags is used to delimit which traits are extracted from the source vs. provided by the user.
- Capability inventory: The skill requests the use of available image or browser tools to inspect external media links found in the reference file.
- Sanitization: The skill relies on structured logic (the reference contract) rather than explicit string sanitization for external content ingestion.
- [SAFE]: The skill uses network operations solely to visit well-known services (X.com) and the project's own evidence platform (goodcase.ai) to inspect visual media, which is consistent with its stated purpose of visual prompt engineering.
Audit Metadata