stitch-design-taste
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied "vibe descriptions" to generate a
DESIGN.mdfile. This generated document is intended to be used as instructions for a downstream AI agent (Google Stitch), creating a surface for indirect prompt injection. - Ingestion points: User's vibe description described in the Atmosphere section of
SKILL.md. - Boundary markers: No explicit delimiters or warnings to ignore embedded instructions were found in the template generation process.
- Capability inventory: The skill's primary output is a design specification markdown file that heavily influences the behavior and output of the Google Stitch AI agent.
- Sanitization: There is no evidence of sanitization, escaping, or validation of the user's input before it is used to synthesize the final design guidelines.
Audit Metadata