horror-suspense-short
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the agent to fetch and inspect media files from
goodcase.aiandmedia.goodcase.ai. These external resources are used as the primary evidence for the 'reference contract' and visual style grounding. - [INDIRECT_PROMPT_INJECTION]: As the skill ingests and summarizes content from external cases, it is theoretically susceptible to indirect prompt injection from untrusted data sources. However, the workflow includes explicit verification steps, requiring the agent to distinguish observed evidence from its own recommendations and to name the specific anchor case used to ensure traceability.
Audit Metadata