design-style-picker
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which could contain hidden instructions designed to influence the agent's behavior.
- Ingestion points: In
SKILL.md(Workflow Step 2) andreferences/selection-playbook.md(Session Pattern), the skill is instructed to inspect rendered UI, screenshots, design tokens, and CSS variables from the project environment. - Capability inventory: The agent has the ability to read these files and subsequently write changes to the codebase during the 'Implement From Selected Images' phase (Workflow Step 6).
- Boundary markers: The instructions lack specific delimiters or warnings to ignore potentially malicious content embedded in the comments of CSS files or the metadata of image assets.
- Sanitization: There is no explicit requirement for the agent to sanitize or validate the external design tokens or screenshots before processing them.
Audit Metadata