seedance-fashion-lookbook
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from
references/cases.md, which contains community-sourced prompt examples. While this serves as an ingestion point for external content, the risk is minimized by the skill's strict workflow and structural constraints. - Ingestion points:
references/cases.md(local reference file). - Boundary markers: Defined 7-block prompt structure and workflow steps.
- Capability inventory: Text generation only; no file-system modification or command execution tools are utilized.
- Sanitization: None.
- [EXTERNAL_DOWNLOADS]: The skill links to external resources for documentation and examples, including GitHub repositories under the
LearnPromptorganization and thegoodcase.aiwebsite. These resources belong to the skill vendor and are used for legitimate informational purposes. - [SAFE]: No patterns of obfuscation, privilege escalation, or unauthorized data access were found. The skill operates within its stated purpose of prompt engineering.
Audit Metadata