react-native-architecture

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: No prompt injection patterns detected. The instructions are purely technical and do not attempt to override agent behavior.
  • [CREDENTIALS_UNSAFE]: The skill demonstrates safe credential handling by using expo-secure-store for authentication tokens and advising against hardcoding secrets in code. Placeholders in configuration files are used appropriately.
  • [EXTERNAL_DOWNLOADS]: The skill references standard Expo and React Native libraries. These are official packages from well-known repositories and registries.
  • [COMMAND_EXECUTION]: Usage of npx and eas commands is restricted to project initialization and build management within the Expo ecosystem.
  • [INDIRECT_PROMPT_INJECTION]: The skill describes patterns for fetching data from external APIs. While this constitutes a data ingestion surface, the provided code follows standard UI patterns with no evidence of unsafe interpolation or exploitable capabilities.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 09:39 AM
Security Audit — agent-trust-hub — react-native-architecture