pmg-visual-assets
Pass
Audited by Gen Agent Trust Hub on Nov 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection by ingesting and processing metadata (e.g., titles, descriptions) from external media repositories.
- Ingestion points: External data enters the agent context during search and retrieval operations from providers like Wikimedia Commons and NASA, as detailed in
SKILL.md. - Boundary markers: The instructions do not specify the use of delimiters or explicit "ignore embedded instructions" warnings when handling retrieved asset metadata.
- Capability inventory: The skill records provenance and technically prepares media files, which includes downloading assets to local directories and creating structured records (
SKILL.md). - Sanitization: There is no evidence of sanitization or character filtering for the text content extracted from external asset metadata before it is stored or displayed.
Audit Metadata