The Agent Skills Directory

[SAFE]: The skill provides technical guidelines for an AI agent to act as a reviewer or advisor for React code. It focuses on optimization patterns such as composition-before-memoization, stale closure detection, and data-fetching race condition handling. No malicious instructions were detected.
[PROMPT_INJECTION]: The instructions do not contain any patterns typical of prompt injection, such as attempts to override system guidelines, bypass safety filters, or switch to unrestricted modes.
[DATA_EXFILTRATION]: There are no commands or code patterns that attempt to access sensitive files (.ssh, .env, .aws) or exfiltrate data to external domains. The mention of 'fetch' is limited to providing code examples for the agent to review in a user's project.
[REMOTE_CODE_EXECUTION]: The skill does not perform any remote package installations or download and execute external scripts. It references well-known libraries like 'react-error-boundary' and 'MUI' for context, but does not execute them as part of the skill logic.
[OBFUSCATION]: No obfuscated content, hidden URLs, or character-encoding tricks (Base64, hex, zero-width characters) were found in the provided files.

react-performance