design-palette
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from local project files (CSS variables and color sources) to normalize design tokens. This creates an indirect prompt injection surface where malicious content within project files could potentially influence agent behavior during the processing workflow. However, this is an inherent functional requirement for design tools and does not represent an active threat.
- Ingestion points: Existing project CSS files, hard-coded color values, and the design-contract.md reference file located via
${CLAUDE_SKILL_DIR}. - Boundary markers: None identified in the workflow instructions.
- Capability inventory: Local file system write access to create or modify CSS theme files (e.g., src/styles/theme.css).
- Sanitization: No explicit sanitization or validation of ingested design data is specified in the instructions.
Audit Metadata