image-to-video
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides technical documentation and curl examples for interacting with the eachlabs API. All referenced domains such as eachlabs.run and eachlabs.ai are vendor-owned assets. No unauthorized file access or credential exfiltration attempts were detected.
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external media through image URLs.
- Ingestion points: image_urls parameter in API requests (SKILL.md).
- Boundary markers: None present in the documentation examples.
- Capability inventory: Media generation via the eachsense/beta model.
- Sanitization: The skill documentation references content policy violation handling as a server-side safety mechanism.
Audit Metadata