expo-project-structure
Fail
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill instructs the agent to execute code from the npm registry using the command
npx --yes submit-expo-feedback@latest. This package is not from a verified organization and uses the@latesttag without a pinned version, which allows for the execution of arbitrary code that could be modified by the package maintainer at any time. - [COMMAND_EXECUTION]: The feedback submission command in
SKILL.mdinterpolates user-provided input ("<actionable feedback>") directly into a shell command. This creates a command injection vulnerability where a user or an external source (via indirect injection) could provide a string containing shell metacharacters (such as;,&&, or|) to execute unauthorized commands on the host machine. - [EXTERNAL_DOWNLOADS]: The skill facilitates the download and execution of external code from the npm registry without integrity checks, posing a supply-chain risk.
Recommendations
- AI detected serious security threats
Audit Metadata