xcode-project-setup

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDYNAMIC_EXECUTIONPROMPT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The instructions utilize strong reinforcement language and specific constraints ('CRITICAL RULES', 'Anti-Ruby Mandate', 'hefty fee') to override default agent behavior and ensure the use of the provided Swift tooling over common alternatives like CocoaPods or Ruby-based scripts.
  • [COMMAND_EXECUTION]: The skill executes environment checks via swift --version and runs the core logic using swift run. These commands are necessary for the skill's primary function of project manipulation.
  • [EXTERNAL_DOWNLOADS]: During the execution of swift run, the Swift Package Manager fetches dependencies defined in Package.swift and Package.resolved, including libraries such as XcodeProj, AEXML, PathKit, and Spectre. These are well-known, widely-used community libraries in the iOS development ecosystem. It also references the official Firebase iOS SDK repository for updates.
  • [DYNAMIC_EXECUTION]: The skill employs swift run to compile and execute local Swift source code located in the scripts/xcode_spm_setup directory at runtime. This is the intended mechanism for performing project modifications using the XcodeProj library.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied data, such as repository URLs and product modules, and interpolates them into command-line arguments for the Swift script. While this represents a vulnerability surface for argument injection, the risk is mitigated by the specific context of local project development and the structured usage patterns defined in the instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 03:56 AM
Security Audit — agent-trust-hub — xcode-project-setup