web-state-mobx
Pass
Audited by Gen Agent Trust Hub on Oct 19, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides standard architectural patterns and educational examples for MobX state management in frontend applications. No malicious patterns were identified in the instruction files or examples.
- [EXTERNAL_DOWNLOADS]: References the well-known and trusted libraries
mobxandmobx-react-litefor client-side state management. The metadata configuration also references a schema hosted on the author's official GitHub repository (agents-inc). - [INDIRECT_PROMPT_INJECTION]: The skill demonstrates patterns for ingesting data from external API clients into reactive state stores, which represents a potential surface for indirect prompt injection if data is later rendered unsafely.
- Ingestion points: Data enters the application state through API responses in
UserStore.fetchUsersandProjectStore.fetchProjectswithinexamples/advanced.md. - Boundary markers: No explicit prompt boundaries or "ignore embedded instructions" warnings are shown in the provided code examples.
- Capability inventory: The skill's capabilities are focused on local state management and do not include dangerous operations such as arbitrary command execution, sensitive file system access, or uncontrolled network exfiltration.
- Sanitization: Data sanitization and validation logic are not addressed in the patterns, as the focus is on reactive state logic and derivation.
Audit Metadata