Color Palette Generator
Pass
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill architecture creates a potential surface for indirect prompt injection by ingesting untrusted external data.
- Ingestion points: Workflows 3 and 4 in
SKILL.mdinvolve retrieving data from external websites viaFirecrawlandPlaywright, as well as processing user-provided images and logos. - Boundary markers: The instructions do not specify any delimiters or safety warnings to prevent the agent from executing instructions potentially embedded in the source websites or image metadata.
- Capability inventory: The skill is configured to use
WebSearch,Firecrawl, andPlaywright, and it generates code exports (CSS/Tailwind configuration), providing a functional path for injected instructions to influence downstream outputs. - Sanitization: There is no evidence of sanitization, filtering, or validation of the content extracted from external URLs or files before it is processed by the agent.
Audit Metadata