icon-composer-app-icon-workflow

Pass

Audited by Gen Agent Trust Hub on Nov 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions and examples for executing the ictool command-line utility located within the Xcode application bundle to export icon previews.
  • Example path: /Applications/Xcode.app/Contents/Applications/Icon Composer.app/Contents/Executables/ictool
  • [COMMAND_EXECUTION]: Includes validation commands intended for development or implementation checks.
  • Commands: uv run scripts/validate_socket_metadata.py and bash plugins/apple-dev-skills/.github/scripts/validate_repo_docs.sh.
  • [EXTERNAL_DOWNLOADS]: References several official documentation and resource URLs from Apple's developer domain for verification of behavior and platform requirements.
  • Sources: developer.apple.com (Icon Composer documentation, Human Interface Guidelines, and WWDC resources).
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface where untrusted data provided by a user (such as design briefs or file names) is ingested and subsequently used in shell commands or GUI automation.
  • Ingestion points: User-supplied design briefs, source asset file paths (interpolated into ictool commands), and layer names (used in the Icon Composer GUI).
  • Boundary markers: None explicitly defined to separate user data from instructions.
  • Capability inventory: Shell command execution (ictool, bash) and GUI automation via Computer Use.
  • Sanitization: No explicit sanitization or validation steps are defined for user-provided strings before they are utilized in commands or the GUI.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 1, 2026, 04:15 PM