explainer-video
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data including scripts, transcripts, and attached files, which could contain instructions intended to influence the agent's behavior.
- Ingestion points: User-provided project states, files, and transcripts are read in SKILL.md (Step 1).
- Boundary markers: The instructions mandate respecting user structure and obtaining acceptance for plan changes, but they do not define explicit delimiters to isolate data from instructions within those inputs.
- Capability inventory: The workflow involves calls to external modules such as widget-forms, voice, and create-motion-graphics to perform UI operations, text-to-speech, and animation authoring.
- Sanitization: The skill lacks specified mechanisms for filtering or escaping user-provided text before processing it for video planning and asset generation.
Audit Metadata