cross-post-video
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted media data which could contain malicious instructions.
- Ingestion points: The
upload_media_from_urlfunction inSKILL.mdStep 1 accepts user-provided public URLs for video processing. - Boundary markers: No delimiters or
ignore instructionswarnings are defined for the imported content. - Capability inventory: The skill has the ability to publish and schedule posts on external accounts using
create_postandschedule_post. - Sanitization: The instructions do not specify any validation or sanitization of the external media content before processing.
Audit Metadata