v0-lander
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill populates UI prompt templates using user-supplied product and brand metadata. Ingestion points: Product name and brand context fields in SKILL.md. Boundary markers: Absent. Capability inventory: Local validation script (scripts/check-output.py) and external component fetching (npx shadcn). Sanitization: None.
- [COMMAND_EXECUTION]: Includes a Python script, scripts/check-output.py, to validate the quality and content of generated deliverables.
- [EXTERNAL_DOWNLOADS]: Recommends fetching UI components via 'npx shadcn', which interacts with the well-known shadcn/ui library.
Audit Metadata