action-continuity-choreography-by-cle9yv
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill directs the agent to access and analyze media files hosted on goodcase.ai and media.goodcase.ai. These are external domains that facilitate the skill's analysis of video prompt examples but reside outside the controlled local environment.
- [INDIRECT_PROMPT_INJECTION]: The workflow involves fetching data from external URLs via browser or media tools. This creates an ingestion point for untrusted content from sites like goodcase.ai and x.com. There is a risk that malicious instructions could be embedded in the external metadata or page content, which the agent might inadvertently follow. The skill lacks explicit sanitization steps or boundary markers to differentiate between example data and executable instructions.
Audit Metadata