skills/expo/skills/expo-upgrade/Gen Agent Trust Hub

expo-upgrade

Pass

Audited by Gen Agent Trust Hub on Oct 22, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to perform various project maintenance tasks using command-line tools such as npx expo install, npx expo-doctor, and npx expo-codemod. It also includes native environment commands like pod install, gradlew clean, and watchman watch-del-all. These are standard operations within the Expo and React Native development lifecycle.
  • [EXTERNAL_DOWNLOADS]: The skill references several remote utilities via npx, including expo-doctor, expo-codemod, and submit-expo-feedback. These resources originate from the vendor's ecosystem ('expo') and are intended for project diagnostics, automated code migration, and feedback submission respectively.
  • [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to read and analyze untrusted project data, including package.json, app.json, and various configuration scripts (metro.config.js, babel.config.js). This creates a surface where malicious content in a project file could attempt to influence the agent's upgrade recommendations, although the specific scope of the skill's instructions limits the potential impact.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 22, 2026, 08:41 PM