redesign
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external codebase content during the audit and scan phases, creating a potential attack surface for instructions embedded in analyzed code. * Ingestion points: The user's codebase is scanned in Step 2 of SKILL.md. * Boundary markers: No specific delimiters or instructions to ignore embedded content are provided. * Capability inventory: Shell command execution via npx (Step 1) and Python script execution (Step 6). * Sanitization: No sanitization or filtering of external content is identified.
- [DYNAMIC_EXECUTION]: The skill executes local Python scripts as part of its verification flow. * Step 6 invokes validate_contrast.py and lint_hardcodes.py from the local .claude/skills/_ux-ui-shared/scripts/ directory.
- [COMMAND_EXECUTION]: The skill uses shell commands to manage UI library components. * Step 1 instructs the agent to use npx shadcn@latest add to install primitives.
- [EXTERNAL_DOWNLOADS]: The skill fetches components from a well-known service using standard tooling. * Downloads UI primitives from the shadcn registry using the npx shadcn tool.
Audit Metadata