vercel-react-best-practices
Warn
Audited by Gen Agent Trust Hub on Apr 5, 2026
Risk Level: MEDIUMPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Deceptive Metadata and Impersonation. The skill's frontmatter and internal documentation (AGENTS.md) explicitly claim the content is maintained by 'Vercel Engineering' and 'vercel', which contradicts the system-provided author identity ('iliagerman'). This misleading attribution may cause agents or users to grant the skill higher trust or execution permissions than appropriate for an unverified third-party contributor.
- [PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill is designed to ingest and process untrusted external data in the form of React components and Next.js logic to provide automated refactoring advice. This creates a surface where malicious instructions embedded in user code could influence the agent's behavior.
- Ingestion points: React components, Next.js pages, and data fetching logic (identified in SKILL.md).
- Boundary markers: Absent; the skill lacks instructions for the agent to distinguish between its own guidelines and potentially malicious instructions within processed user files.
- Capability inventory: The skill is intended to guide code generation, review, and automated refactoring, which typically implies the executing agent has permissions to modify local files.
- Sanitization: No input validation or filtering of the code being reviewed is specified.
Audit Metadata