skills/expo/skills/expo-web-to-native/Gen Agent Trust Hub

expo-web-to-native

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Instructions throughout the skill and its references guide the user to execute shell commands for project initialization, running the development server, and using verification tools. This includes npx expo start, xcrun simctl for iOS simulators, and adb for Android emulators.
  • [EXTERNAL_DOWNLOADS]: The skill recommends installing external packages from the NPM registry to facilitate automated testing and verification. These include agent-browser (a Vercel Labs tool) and @swmansion/argent (an ecosystem partner tool).
  • [REMOTE_CODE_EXECUTION]: The feedback mechanism utilizes npx --yes to download and run the submit-expo-feedback package directly from the NPM registry. This command fetches and executes the latest version of the vendor's feedback utility.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze an existing React codebase to formulate a migration strategy. This involves the agent reading and interpreting untrusted source code, which could potentially contain instructions aimed at influencing the agent's behavior.
  • Ingestion points: User-provided web React codebase (read in SKILL.md Step 1).
  • Boundary markers: The skill does not define specific delimiters or warnings for the content it processes.
  • Capability inventory: The agent has access to shell commands and package managers (npm, npx) as part of its development environment.
  • Sanitization: There is no explicit sanitization of the code being analyzed.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 05:02 AM
Security Audit — agent-trust-hub — expo-web-to-native