skills/simota/agent-skills/snap/Gen Agent Trust Hub

snap

Pass

Audited by Gen Agent Trust Hub on Aug 13, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses standard command-line utilities for iOS development, including xcodebuild for test execution and xcrun simctl for simulator management and status bar manipulation. These are well-documented tools for the specified domain.- [PRIVILEGE_ESCALATION]: The skill documents the use of sudo xcode-select within CI environment scripts. This is a routine administrative command required to set the active developer directory on shared CI runners.- [INDIRECT_PROMPT_INJECTION]: The skill provides patterns for processing iOS application UI trees, which constitutes an indirect prompt injection surface.
  • Ingestion points: The skill interacts with the application's UI hierarchy via XCUIApplication and Accessibility Inspector.
  • Boundary markers: Not explicitly provided, but the skill mandates the use of static accessibilityIdentifier paths (e.g., screen.section.element) which limits the impact of dynamic UI content.
  • Capability inventory: Includes file access for .xcresult bundles, execution of build commands, and potential network uploads to device farms.
  • Sanitization: The instructions direct the agent to refactor volatile recorded code into structured Screen Objects using stable identifiers, reducing reliance on attacker-controlled element labels.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 13, 2026, 09:28 AM
Security Audit — agent-trust-hub — snap