ai-product-video-ad
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill allows the agent to process external image content provided via URLs (image_url). This creates a surface where instructions hidden within the image (e.g., via OCR-based triggers or steganography) could potentially influence the agent's actions.
- Ingestion points: SKILL.md (With Reference Image section).
- Boundary markers: None provided to delimit untrusted image content from system instructions.
- Capability inventory: The skill utilizes the Bash tool (curl) and WebFetch to interact with external APIs.
- Sanitization: No sanitization or content validation for external images is described.
Audit Metadata