aesthetic
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection (LOW): The skill's primary workflow involves analyzing untrusted content from external design websites. This provides an entry point for malicious instructions embedded in those sites. Evidence: Ingestion points in references/workflows.md (Workflow 1); Boundary markers are absent; Capability inventory includes chrome-devtools and bash; Sanitization is absent.
- Command Execution (LOW): The skill suggests using the gallery-dl CLI tool via bash to download images. While standard for the task, this exposes a command execution surface if URLs are not handled securely. Evidence: references/design-resources.md.
- External Downloads (LOW): The skill integrates with multiple external design platforms to retrieve media and design patterns.
Audit Metadata