swift-refactor

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is composed entirely of markdown documentation and Swift/SwiftUI code examples intended for architectural guidance. A thorough analysis across all 11 threat categories (including prompt injection, data exfiltration, and persistence) revealed no malicious intent or suspicious behavior.
  • [EXTERNAL_DOWNLOADS]: The skill neutrally references the 'ordo-one/equatable' Swift Package Manager dependency from its official GitHub repository to enable specific macro functionality mentioned in the performance guidelines. All other external links point to established developer documentation and well-known engineering blogs (e.g., Apple Developer, Airbnb Engineering).
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing user-provided code for refactoring. The instructions focus strictly on architectural transformations and modernization patterns, maintaining a safe posture for processing external source code within the agent context.
  • [COMMAND_EXECUTION]: The skill does not contain any executable scripts or shell commands. All provided code snippets are intended for manual or agent-assisted refactoring of iOS application source files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 06:25 PM
Security Audit — agent-trust-hub — swift-refactor