ios-device-automation

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The prompt explicitly shows insecure examples that embed secrets directly into CLI commands (e.g., an act prompt that fills a password 'pass123'), which would require the LLM to include secret values verbatim in generated command text, even though env-var configuration is recommended elsewhere.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill repeatedly instructs running "npx -y @midscene/ios@1" (which fetches and executes remote code from the npm registry at runtime) as a required dependency to control the device, so it downloads and runs external code during execution.