expo-ui

Fail

Audited by Gen Agent Trust Hub on Oct 5, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill recommends executing npx --yes submit-expo-feedback@latest to submit actionable feedback. This command downloads and runs a package from the public NPM registry that is not part of the official Expo organization namespace, posing a significant risk of executing untrusted or potentially malicious code.\n- [EXTERNAL_DOWNLOADS]: The skill uses npx expo install @expo/ui to install necessary dependencies. This is a standard and safe operation within the Expo and React Native development environment, targeting well-known packages.\n- [COMMAND_EXECUTION]: A local utility script, scripts/list-components.js, is authorized to run via the Bash tool. The script parses local TypeScript definition files in the node_modules directory to identify available components and modifiers for the agent.\n- [INDIRECT_PROMPT_INJECTION]: The documentation advises the agent to perform research on external platforms like Jetpack Compose and Material Design 3. Ingesting content from these external sources represents a potential attack surface where malicious instructions could be introduced into the agent's context during the research process.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Oct 5, 2026, 09:27 PM