The Agent Skills Directory

[Overall Assessment] (SAFE): The skill is purely informational and consists of static code snippets for React state management using the Jotai library.
[Category 4: Unverifiable Dependencies] (INFO): References standard, widely-used Node.js packages (jotai, jotai-tanstack-query, jotai-devtools). No suspicious installation scripts or remote execution patterns were found.
[Category 2: Data Exposure & Exfiltration] (INFO): Includes standard web fetch requests to local API endpoints (e.g., /api/user). No sensitive data access or exfiltration to untrusted domains is present.
[Category 8: Indirect Prompt Injection] (INFO): The skill is static documentation and does not ingest untrusted data for processing or agent decision-making. There is no attack surface for prompt injection.

jotai-patterns