video-style-prompt-advisor
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to translate untrusted user input, such as video project requirements and mood board descriptions, into structured 'style preset' prompts intended for downstream image generation models like GPT image2. This creates an architectural surface for indirect prompt injection.\n- Ingestion points: User project requirements, including题材 (subject matter) and reference directions, are ingested in Step 1 of the working process defined in SKILL.md.\n- Boundary markers: The instructions do not define clear delimiters or specific directives to ignore potentially malicious instructions embedded within the user's project description when generating the final prompt.\n- Capability inventory: The skill is designed to produce natural language instructions that are explicitly meant to control the visual output logic of other AI models.\n- Sanitization: There is no evidence of input validation, escaping, or filtering to prevent adversarial content in the user's request from being passed into the generated style presets.
Audit Metadata