Make Podcast
Pass
Audited by Gen Agent Trust Hub on Aug 13, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user data in the form of a
scriptcontaining dialogue lines (line) which are then processed by a media generation pipeline. - Ingestion points: The
scriptarray in thePOST https://api.agent-media.ai/v1/skills/make_podcast/runrequest, specifically thelinefields for speakers A and B. - Boundary markers: None identified. There are no instructions to the model to escape or ignore malicious instructions embedded within the dialogue lines.
- Capability inventory: The skill performs network operations via HTTP POST and GET requests to
api.agent-media.ai. - Sanitization: No explicit sanitization or validation of the
linecontent is mentioned before it is sent to the external service.
Audit Metadata