jaredlander/freshbooks-speed
GitHubcss
Expert-level CSS development with modern features and best practices. Use when asked to (1) write or debug CSS, (2) implement layouts with flexbox, grid, or container queries, (3) create animations and transitions, (4) optimize CSS performance, (5) work with CSS preprocessors or CSS-in-JS, (6) implement responsive design, or when phrases like "style", "CSS", "stylesheet", "design", "layout", "animation" appear.
javascript-ember
Expert-level Ember.js development. Use when asked to (1) write Ember.js applications with components, services, routes, or controllers, (2) implement Ember Data models and adapters, (3) work with Ember Octane patterns (Glimmer components, tracked properties, modifiers), (4) optimize Ember application performance, (5) write Ember tests with QUnit or testing-library, or when phrases like "Ember component", "Ember route", "Glimmer", "tracked property", "Ember addon" appear.
html
Write semantic, accessible, performant HTML with modern best practices. Use when asked to (1) create HTML pages or documents, (2) write semantic markup, (3) improve accessibility, (4) optimize HTML structure and performance, (5) implement forms, tables, or complex layouts, or when phrases like "HTML page", "web page", "markup", "semantic HTML", "accessibility" appear.
javascript-react
Expert-level JavaScript and React development. Use when asked to (1) write JavaScript code requiring advanced patterns like closures, proxies, generators, or async iterators, (2) build React applications with hooks, context, suspense, or server components, (3) optimize JavaScript/React performance, (4) implement complex state management, (5) write TypeScript with advanced type patterns, or when phrases like "React component", "JavaScript function", "TypeScript", "hooks", "state management", "frontend", "web app" appear.
unit-testing
单元测试编写指南，涵盖 JUnit5/MockK 使用、测试命名规范、Mock 技巧、测试覆盖率要求、TDD 实践。当用户编写单元测试、Mock 依赖、提高测试覆盖率或进行测试驱动开发时使用。
code-review
Use when receiving code review feedback (especially if unclear or technically questionable), when completing tasks or major features requiring review before proceeding, or before making any completion/success claims. Covers three practices - receiving feedback with technical rigor over performative agreement, requesting reviews via code-reviewer subagent, and verification gates requiring evidence before any status claims. Essential for subagent-driven development, pull requests, and preventing false completion claims.