v0-lander
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill instructs users to run
npx shadcn@latest add [component-url]to import UI components. This command fetches and executes code from the v0.dev platform, which is a well-known service provided by Vercel for generative UI. - [INDIRECT_PROMPT_INJECTION]: The skill creates prompts for generative AI by interpolating user-provided data such as brand context and product descriptions.
- Ingestion points: Phase 1 and 2 in
SKILL.mdtake user inputs for branding and product details. - Boundary markers: The templates use brackets for placeholders but do not include instructions for the AI to ignore potentially malicious embedded content.
- Capability inventory: The skill uses a local script
scripts/check-output.pyfor content validation, which has limited file system access. - Sanitization: No escaping or validation is performed on the user inputs before they are placed into the prompt templates.
Audit Metadata