ai-cinematic-video-director
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEMETADATA_POISONINGINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides a filmmaking methodology for generating AI video prompts and does not include any executable code, file operations, or network communication.
- [METADATA_POISONING]: The stats.json metadata file includes a future date in the updatedAt field (2026-06-16). This is anomalous but does not pose a functional security threat.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user input to generate cinematic prompts (Ingestion: User requests). Boundary markers and sanitization are not provided. However, the capability inventory reveals no access to tools, files, or external networks, rendering the injection surface safe.
Audit Metadata