The Agent Skills Directory

[SAFE]: The skill consists entirely of documentation and React code examples. No executable scripts, command execution, network exfiltration, or unauthorized file access patterns were identified.
[METADATA_POISONING]: The author field in SKILL.md is listed as 'vercel', identifying the original source of the guidelines, while the publisher is 'tyler-r-kendrick'. This discrepancy is common for documentation-focused skills and does not represent a security risk.
[PROMPT_INJECTION]: The skill uses high-impact terminology such as 'Impact: CRITICAL' to label architectural principles in AGENTS.md and related files. These labels are contextual to component design and do not attempt to override agent safety protocols.

vercel-composition-patterns