NYC
skills/smithery/ai/build-iphone-apps/Gen Agent Trust Hub

build-iphone-apps

Fail

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: HIGHCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION] (HIGH): The verification loop executes shell commands including xcodebuild -scheme AppName and xcrun simctl launch booted com.company.AppName. These parameters are derived from user-provided project metadata. Without explicit sanitization or strict boundary markers, this provides a direct path for shell command injection via specially crafted project or scheme names.
  • [REMOTE_CODE_EXECUTION] (HIGH): The skill's primary function is to generate and compile Swift code based on untrusted user descriptions. This represents a 'Tier: HIGH' Indirect Prompt Injection surface where the agent acts as an automated 'execute/write' component for external content, lacking necessary sanitization or human-in-the-loop validation for the generated logic.
  • [EXTERNAL_DOWNLOADS] (MEDIUM): The verification logic pipes build output to a tool named xcsift (xcodebuild ... | xcsift). This utility is not a standard component of the Xcode command-line tools or macOS. Its source, integrity, and safety are unverifiable, posing a supply-chain risk if the environment requires manual installation of this dependency.
  • [CREDENTIALS_UNSAFE] (LOW): Automated scans flagged com.app as a phishing-related domain. This string is used as a placeholder bundle identifier in the xcrun simctl command. While intended as a placeholder, its use in executable strings matches patterns associated with malicious URL delivery.
Recommendations
  • AI detected serious security threats
  • Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 16, 2026, 08:24 AM