skills/expo/skills/eas-update/Gen Agent Trust Hub

eas-update

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses official CLI tools like expo and eas-cli to perform update configuration, publishing, and debugging tasks.
  • [EXTERNAL_DOWNLOADS]: Fetches and runs official Expo packages (expo-updates, eas-cli, submit-expo-feedback) from the NPM registry via npx. These are recognized vendor resources and are safe within this context.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: The skill inspects project configuration files such as package.json, eas.json, and native app configs.
  • Boundary markers: None defined for reading project files; the skill relies on structural data.
  • Capability inventory: Uses subprocess execution for EAS tasks and feedback submission.
  • Sanitization: Standard parameter handling by vendor-provided CLI tools is utilized.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 3, 2026, 07:39 PM
Security Audit — agent-trust-hub — eas-update