epic-fantasy-scifi-spectacle-by-uaxh44

Pass

Audited by Gen Agent Trust Hub on Sep 22, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references external URLs in references/cases.md including domains like goodcase.ai and x.com. The instructions in SKILL.md direct the agent to use browser or media tools to inspect these external resources to ensure visual fidelity during the prompt generation process.
  • [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting data from external URLs. Ingestion points: External links to goodcase.ai and x.com defined in references/cases.md. Boundary markers: The workflow requires a structured 'Reference Contract' (Preserve, Replace, Avoid) and a multi-step verification process to maintain control over the generated output. Capability inventory: The skill uses browser, image, and video tools to analyze external content. Sanitization: There is no explicit sanitization of external content, but the impact is minimized by the generative nature of the task and the requirement to map three structural traits back to the source evidence.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 22, 2026, 09:04 AM