The Agent Skills Directory

[SAFE]: The skill documents standard architectural patterns for React state management using established libraries.
[SAFE]: No suspicious remote code execution, obfuscation, or data exfiltration patterns were detected.
[SAFE]: References to external repositories like @lukemorales/query-key-factory and official TanStack documentation are well-known resources in the frontend ecosystem.
[INDIRECT_PROMPT_INJECTION]: The skill describes patterns for ingesting data from external APIs and local storage. While this represents a potential surface for indirect injection (e.g., malicious content in a database record influencing the UI), the skill does not implement unsafe execution or interpolation logic that would escalate this beyond a baseline risk inherent to web development.
Ingestion points: queryFn implementations (fetching server data, reading localStorage).
Boundary markers: Not explicitly mentioned, as this is a library usage guide rather than a implementation of a specific agentic pipeline.
Capability inventory: None. The skill only provides React hook patterns.
Sanitization: Not explicitly discussed; follows standard React rendering safety (automatic escaping of JSX).

tanstack-query-v5