vibe-coding-design-guard
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted external data which could contain instructions intended to manipulate the agent's review outcome.
- Ingestion points: The skill reads
TECH_STACK.md,DESIGN_SYSTEM.md, and various source code files (JSX, TSX, Vue, Svelte, etc.) to perform design token validation. - Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are defined to isolate the content of the analyzed files from the skill's operational directives.
- Capability inventory: The skill is designed to block or allow code generation and commits based on its analysis, giving it control over the development workflow.
- Sanitization: The skill lacks explicit sanitization or validation for the text extracted from design systems and components beyond simple regex-style token matching.
Audit Metadata