About this project

Mobile Engineering Agents is a drop-in knowledge system designed to enhance the engineering judgment of AI coding agents such as Claude Code, Cursor, Windsurf, Gemini CLI, and Aider. It focuses on transforming AI-generated code from merely plausible to production-grade by providing a structured framework of architecture, security, and testing standards. The toolkit is primarily focused on iOS (Swift/SwiftUI), with support for UIKit, Android (Kotlin), and Flutter. It organizes its knowledge into several key components: - Agents: 14 loadable expert roles including System Design, Security, and Performance experts. - Skills: 31 deep-dive topics covering authentication, websockets, and caching. - Workflows: 11 step-by-step procedures for building features or integrating APIs. - Checklists: 8 objective review gates for automated quality assurance. - Standards & Architecture: Non-negotiable coding rules and reference designs with Mermaid diagrams. Users can integrate the toolkit by cloning it into their project and wiring it to their specific AI tool's configuration file. The system operates via a classification and routing mechanism: it identifies the task, routes it to the appropriate expert agent, chains the task through implementation and hardening tiers, and concludes with a self-review against a checklist.