The Agent Skills Directory

[SAFE]: No security issues detected. The skill functions as a static reference for frontend state management patterns.
[SAFE]: The skill's configuration includes disable-model-invocation: true, which is a strong security measure that prevents the model from invoking tools autonomously while this skill is in context.
[SAFE]: Code examples in references/middleware-composition.md demonstrate security best practices by providing a serialization replacer to redact sensitive fields like 'password' and 'token' from being exposed to browser devtools.
[SAFE]: All identified dependencies (e.g., immer, idb-keyval) are well-known, industry-standard libraries for the React/Zustand ecosystem.
[SAFE]: No evidence of prompt injection, obfuscation, or unauthorized network operations was found across the documentation and script templates.

zustand-patterns