ui-designer
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a shell command (
find . -name "package.json" -exec grep -l "react" {} \;) to automatically detect the presence of a React project within the current workspace. - [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface as it ingests untrusted user data (UI screenshots and project description files) and interpolates this content into prompts for subagents.
- Ingestion points: User-provided reference images directory (Step 2) and project idea file (Step 3) defined in
SKILL.md. - Boundary markers: The templates in
assets/use XML-like tags (e.g.,<context>,<app-overview>) to delimit interpolated content. - Capability inventory: The skill has the ability to execute shell commands (
find), write files to thedocuments/directory, and generate/save React component files (.jsx). - Sanitization: No explicit sanitization or filtering of the user-provided text/metadata is performed before interpolation into subagent contexts.
Audit Metadata