sd-jieshuo-engine
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill consists entirely of instructional Markdown files. It defines creative standards, narrative structures, and visual consistency rules without including any executable scripts or automation logic.
- [SAFE]: The instructions are focused on domain-specific workflows for short-form video production (Track 3: Explainer/Narrative). It specifies parameters for voice-over scripts, historical accuracy markers, and visual style consistency.
- [INDIRECT_PROMPT_INJECTION]: The skill defines how to process user requests for specific video topics. It mitigates potential injection risks by providing a rigid structural framework (hooks, golden quotes, and source verification tiers) that the AI agent must follow when generating content. The mentioned 'C14 exemption' refers to domain-specific validation logic within the platform's internal quality control scripts and does not represent a safety guideline bypass.
Audit Metadata