bootstrap-ios

Pass

Audited by Gen Agent Trust Hub on Aug 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill includes a local shell script scripts/bootstrap-ios-skills.sh which the user or agent can run to automate environment setup. The script executes commands such as npx skills add to register new skills and npx xcodebuildmcp init to initialize development tools.
  • [EXTERNAL_DOWNLOADS]: The bootstrap script identifies and fetches multiple iOS-related agent skills from public GitHub repositories. The sources include well-known community members (e.g., twostraws, AvdLee) and established technology services (e.g., Sentry's getsentry/XcodeBuildMCP). These references are documented and the script includes a dry-run mode for transparency.
  • [PROMPT_INJECTION]: The skill's primary function involves analyzing local Apple-platform project files (such as *.xcodeproj, Package.swift, and *.swift). This creates a surface for indirect prompt injection, where malicious content embedded in project metadata or code comments could attempt to influence the agent's behavior. However, the skill maintains a conservative posture and recommends verification steps to mitigate unintended actions.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 13, 2026, 12:40 PM
Security Audit — agent-trust-hub — bootstrap-ios