The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains deceptive metadata in the SKILL.md frontmatter, identifying the author as 'vercel'. This contradicts the actual author identity, 'Nairon-AI'. Impersonating a well-known, trusted technology vendor is a misleading practice that could influence a user's or agent's trust in the skill's safety.
[SAFE]: The code examples and React composition patterns provided in AGENTS.md and the rules/ directory are benign documentation for architectural improvement. They do not involve any network operations, sensitive file access, or command execution.
[SAFE]: External references are restricted to official documentation on react.dev, which is a recognized and safe source.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external source code to perform refactoring tasks, which creates a surface for indirect prompt injection.
Ingestion points: Source code files provided to the agent for architectural review and refactoring.
Boundary markers: The instructions do not define specific delimiters or warnings to ignore instructions that might be embedded in the code comments of the files being processed.
Capability inventory: The skill facilitates code modification and structural generation.
Sanitization: No validation or sanitization logic is prescribed for the input data.

vercel-composition-patterns