horror-chase
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read from external sources, including "existing project assets" and the "gameplay-experience-study archive".
- Ingestion points: The skill explicitly mentions pointing to existing assets and reading specific IDs from an archive to influence the design output.
- Boundary markers: The provided instructions do not include specific delimiters or instructions to ignore potential commands embedded within these external project assets or study data.
- Capability inventory: The skill is designed to generate text-based designs, map plates, and video plans (via
film-action-combat). It does not appear to have direct file-write or network-send capabilities itself, but it influences downstream generation processes. - Sanitization: No sanitization or validation logic is specified for the ingested content, which could allow maliciously crafted project assets to influence the agent's behavior during the design process.
Audit Metadata