react-architecture
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides structural guidance and architectural rules for frontend development. The content is focused on code quality and maintainability without introducing security risks or malicious behaviors.- [EXTERNAL_DOWNLOADS]: The skill refers to and fetches technical documentation from the official TanStack website to ensure the agent uses correct API versions and implementation patterns for the project's stack.- [COMMAND_EXECUTION]: Includes instructions and configuration for running
dependency-cruiser, a developer tool used to verify that code follows the defined architectural boundaries and dependency rules.- [INDIRECT_PROMPT_INJECTION]: The skill identifies and manages surfaces where untrusted data enters the agent context. - Ingestion points:
validateSearchinroutes/orders/index.tsx(URL params) andunwrapinlib/api-client.ts(API responses). - Boundary markers: Implements explicit Zod schemas (
OrderListSearchSchema,ErrorBodySchema) to define acceptable data structures. - Capability inventory: The skill is limited to UI component rendering and navigation; no dangerous capabilities like arbitrary shell execution or local file writing were identified.
- Sanitization: Employs mandatory Zod parsing to validate and sanitize all external data at the point of ingestion.
Audit Metadata