skills/midudev/autoskills/clerk-expo/Gen Agent Trust Hub

clerk-expo

Pass

Audited by Gen Agent Trust Hub on Nov 8, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides clear and secure instructions for implementing authentication in Expo applications. It includes multiple 'Execution Gates' that enforce security checks, such as verifying environment variable prefixes and ensuring the Native API is enabled in the dashboard.
  • [CREDENTIALS_SAFE]: Instructions explicitly forbid hardcoding keys and require reading them from environment variables with the EXPO_PUBLIC_ prefix. It also mandates the use of expo-secure-store for session persistence instead of unencrypted alternatives like AsyncStorage.
  • [EXTERNAL_DOWNLOADS]: All suggested dependencies are official, well-known packages from the Clerk and Expo ecosystems (e.g., @clerk/expo, expo-secure-store). Installation is performed using the standard npx expo install command, which ensures version compatibility.
  • [COMMAND_EXECUTION]: Shell commands are restricted to standard development workflows such as running the app (npx expo run:ios), performing type checks (npx tsc), and project initialization (npx clerk init).
  • [INDIRECT_PROMPT_INJECTION]: While the skill reads project configuration and local SDK files to verify versions, it does not process untrusted external data in a way that would allow for prompt injection. The scope is limited to the local development environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 8, 2026, 10:44 PM