expo-dev-client

Warn

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: MEDIUMREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill includes a command npx --yes submit-expo-feedback@latest that downloads and runs code from the NPM registry. The use of the @latest tag means the skill does not verify the version or integrity of the package at runtime, which could result in the execution of malicious updates if the package is compromised.\n- [COMMAND_EXECUTION]: The feedback command npx --yes submit-expo-feedback@latest --category skills --subject "expo-dev-client" "<actionable feedback>" interpolates user-controlled text directly into a shell command string. Without explicit sanitization or escaping, this could allow for arbitrary shell command injection if the input contains shell metacharacters.\n- [EXTERNAL_DOWNLOADS]: The skill references the submit-expo-feedback package from the public NPM registry. This package is not an official tool from the primary technology vendor (Expo) and is fetched dynamically.\n- [INDIRECT_PROMPT_INJECTION]: The feedback ingestion point creates a potential for indirect injection attacks.\n
  • Ingestion points: The <actionable feedback> parameter within the feedback submission command in SKILL.md.\n
  • Boundary markers: No boundary markers or 'ignore embedded instructions' warnings are present to delimit user-supplied content.\n
  • Capability inventory: The skill possesses the ability to execute shell commands (e.g., npx, eas, adb, xcrun, tar).\n
  • Sanitization: No sanitization or validation of the user-provided string is performed before it is interpolated into the shell command execution path.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 14, 2026, 01:43 PM
Security Audit — agent-trust-hub — expo-dev-client