vercel-composition-patterns

Warn

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: MEDIUMPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill metadata in SKILL.md and metadata.json falsely identifies the author as 'vercel'. This is deceptive because the actual author is 'lingxling'. Impersonating a trusted vendor can mislead users into assuming higher safety and quality standards than are verified for the actual author.
  • [NO_CODE]: The skill consists exclusively of documentation and code examples in Markdown format, with no executable scripts, binaries, or automated tools.
  • [SAFE]: The technical content and external links to react.dev are legitimate and relate to standard React development practices, with no malicious code or instructions detected.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 4, 2026, 10:21 AM