forge_implement-feature
Use when 需要实现一个新 Feature，从 Schema 定义到数据层、服务层、状态层、UI 层再到最终质量检查，串联多个原子动词 Skill 完成宽线性工作流。触发词：实现功能、开发功能、新增 feature。
forge_implement-split
Use when 需要将大型组件或模块按最佳实践拆分为多个独立文件，系统性地执行代码拆分与重构，确保拆分后各模块结构符合规范。触发词：拆分组件、实现文件拆分、代码模块拆分重构、component-split。
forge_implement-trpc-query
Use when 需要实现 tRPC 按需查询（on-demand query）模式，确保查询逻辑和数据获取方式符合项目 tRPC 规范和最佳实践。触发词：tRPC按需查询、实现tRPC查询、on-demand查询模式。
forge_integrate-better-auth
Use when 需要把 Better Auth 集成到项目，涵盖安装配置、数据库集成、Provider 设置和认证流程验证等完整步骤。触发词：集成 better-auth、接入 better-auth、better-auth 集成。
forge_integrate-zod-env
Use when 需要配置类型安全的环境变量管理模块，基于 .env.example 使用 Zod 生成 env schema 和标准化管理工具。触发词：zod 环境变量、env 类型安全、env schema。
forge_introspect-best-practice
Must follow when 深度反思已定义的 best-practice，识别盲区、过度设计、隐藏假设和未覆盖场景，并提出可验证的规范改进建议。触发词：反思最佳实践、深挖规范盲区、自检 skill。
forge_no-re-export-best-practice
Must follow when 审查或编写非 index 文件中的 import/export，禁止在非桶导出文件中对外部模块做 re-export（间接转发），每个消费文件应直接从来源 import。
forge_one-component-per-file-best-practice
Must follow when 检查或重构 React/Vue 组件文件，强制每个文件只包含一个组件，不允许多组件共存于同一文件；支持 TSX/JSX/Vue。
forge_page-best-practice
Must follow when 创建或审查前端页面结构，确保遵循 Anatomy 规范，正确分离 Wrapper、Content 和 Optional Store 模块。
forge_refactor-classname
Use when 需要检查或转换 React/Vue 文件中的 className 模板字符串为 cn 函数调用，支持递归扫描文件夹、详细报告所有位置。触发词：优化className写法、重构模板字符串、检查className规范。
forge_refactor-ui-components
Use when 需要将代码中的原生 HTML 元素（如 button、input、select 等）自动重构为项目组件库组件，确保符合组件库规范并保持功能不变。触发词：重构 UI 组件、替换原生元素、迁移到组件库。
forge_refine-trpc-best-practice
Must follow when 在 React 组件中进行数据获取，确保通过 Refine hooks（useList/useOne 等）经由 DataProvider 访问数据，禁止直接调用 trpc 客户端。
forge_remove-comments
Use when 需要批量删除代码文件中的注释，支持 Python、JS、TS、TSX、Java、C/C++、Rust、Go、HTML 等主流编程语言。触发词：删除注释、清除代码注释、移除注释行、批量去注释。
forge_repository-best-practice
Must follow when 创建或重构 Repository 层，确保数据访问模式、方法命名、返回类型和接口定义符合项目规范。
forge_schema-best-practice
Must follow when 创建或重构 Zod Schema，确保目录、命名、类型推导和边界校验符合 Daedalus 规范。
forge_service-best-practice
Must follow when 创建或重构 Service 层，基于 tRPC + Service + DAO 架构确保依赖注入、错误处理和业务逻辑分层符合规范。
forge_skill-best-practice
Must follow when 创建或修改 Skill 后执行质量验证，涵盖命名、目录结构、元数据完整性、临时文件清理和依赖格式共 16 项检查。
forge_sample-best-practice
Must follow when 作为示范 Skill 用于说明 Skill 最佳实践结构，包含最小化 SKILL.md、完整 references/ 和 best-practice-examples/ 目录。触发词：示范skill、skill结构示例、skill模板参考。
forge_store-best-practice
Must follow when 使用 Zustand 创建或重构状态管理 Store，确保遵循 slice 模式、Provider 设置和类型安全规范。
forge_storybook-best-practice
Must follow when 创建或维护 Storybook Stories，确保组件文档命名、参数定义和装饰器配置符合项目 Storybook 编写规范。
forge_subagent-driven-development
Use when 需要把已批准的实现计划拆成独立任务并分派给子代理执行，通过规格一致性审查与代码质量审查双阶段把关，确保并行实现结果与设计、计划和验收标准保持一致。触发词：子代理开发、分派任务、并行实现、subagent。
forge_svg-icon-best-practice
Must follow when 管理或新增 React TypeScript 项目中的 SVG 图标组件，确保命名、封装方式和导出规范遵循项目标准规范。
forge_test-driven-development
Use when 需要以测试先行方式实现功能、修复 bug 或重构行为，严格执行 Red-Green-Refactor 循环，先写失败测试再写最小生产代码并完成重构验证。触发词：TDD、测试先行、红绿重构、先写测试。
forge_ui-components-best-practice
Must follow when 创建或审查 UI 组件代码，确保在标准场景下使用项目组件库组件（如 shadcn/ui），禁止直接使用原生 HTML 元素。
forge_use-store-not-props-best-practice
Must follow when 设计组件数据流——优先通过 Store 访问全局状态，不得通过 Props 层层传递；适用于代码审查和新功能设计阶段。
forge_writing-plans
Use when 需要将已批准的设计规格拆解为可执行实现计划，按最小任务粒度给出文件路径、完整代码、验证命令、依赖上下文和完成标准，便于低上下文执行者稳定落地实现。触发词：写计划、实现计划、任务拆解、plan。
forge_zod-infer-type-best-practice
Must follow when 项目中存在 Zod Schema 定义时，禁止重复声明可推导的业务类型；可推导类型使用 z.infer，跨层 contract 保持独立职责。
maestro-bundle
Route work into the right maestro tier and drive the SPEC/NOTES/VERIFY bundle lifecycle - open, resume, close, recall.
maestro-coach
Decision support and grounded teaching for the human. Invoke when the user signals a message did not land ("wait, what?", "I don't understand", "explain it like I'm five", "which is better?", "I'm not sure") about a question or fork they have been asked, or when they hand over a locked decision, a lesson, or a bundle and want to learn the concept behind it. Also triggers on Vietnamese phrasings such as giải thích chi tiết, giải thích rõ hơn, rõ hơn, tôi không hiểu, cái nào tốt hơn.
maestro-council
Lead-only council for a hard-to-reverse fork - a neutral brief, sealed independent seats, one premise verifier on unanimity, bounded verifiers, one cross-examination round, a draft-verdict audit, and one binding verdict recorded as a decision with its dissent. Never inside a seat.
maestro-design
Resolve material unknowns blocking the next authorized slice, using research, grilling, prototypes, models, or wayfinding. Record durable decisions and apply the shared workflow tier rule.
maestro-diagnose
Diagnosis-only investigation of a failure with unknown cause - reproduce, localize, reduce, deliver root cause with evidence. Read-only; an explicit fix request routes by target and tier instead, never here. Also triggers on Vietnamese phrasings such as root cause là ở đâu, nguyên nhân là gì, tại sao nó không.
maestro-explore
Answer an evidence question without touching production code - research external facts against primary sources, build or iterate a disposable prototype (its own bugfixes included), or baseline current repository behavior before deciding or implementing. Use when the question is answerable by reading, building, or measuring with no user decision required; unsettled decisions belong to maestro-design. Prototypes are throwaway. Also triggers on Vietnamese phrasings such as làm 1 vòng research xem, research xem.
maestro-graph
Drive a pre-known multi-agent path as a maestro graph - run it by name or from a file you just wrote, pull each agent node with graph next, spawn it as a sub-agent under its maestro-<profile> definition, hand the result back with graph result, repeat until the verdict. A sub-agent node is read-only scouting; a node that writes files or is accepted is opened as a pane instead. Author a new graph from the reference when no preset fits.
maestro-improve
Turn filed lessons into the smallest doctrine edit. Use when a Lead assigns improvement work for one target: group pending lessons, make one evidence-linked commit per target, return it for independent challenge, and process each accepted or rejected lesson without deleting history.
maestro-questionnaire
Turn a decision the user cannot fully answer into a Markdown questionnaire for the one person who can - filled in async, or together over a meeting. Also triggers on Vietnamese phrasings such as phỏng vấn đi, phỏng vấn tôi, interview tôi.
maestro-verify
Verify and close - cross-check coverage, run the VERIFY table, deliver the verdict, harvest durable lessons into decisions, close the bundle, and never claim remote state from local evidence.
maestro-work
Implement or fix one authorized unit with minimal edits and sufficient evidence. Reuse existing checks and add tests only for concrete uncovered behavior or risk.
collaborating-with-claude
Use the Claude Code CLI to consult Claude and delegate coding tasks for prototyping, debugging, and code review. Supports multi-turn sessions via SESSION_ID. Optimized for low-token, file/line-based handoff.
collaborating-with-gemini
Use the Gemini CLI to consult Gemini and delegate coding tasks for prototyping, debugging, and code review. Supports multi-turn sessions via SESSION_ID. Optimized for low-token, file/line-based handoff.
latex-rhythm-refiner
>
arxiv-paper-writer
>
laravel-upgrade
Upgrade Laravel applications one major version at a time (9→10, 10→11, 11→12). Use when user wants to upgrade their Laravel framework version. Auto-detects current version from composer.json, identifies breaking changes, and applies necessary code fixes.
peekabo
Use when working with the Peekaboo CLI or MCP server to capture macOS screens, inspect UI elements, and automate GUI interactions (see/click/type/scroll/hotkey/window/menu/dock/space), or when troubleshooting Peekaboo permissions, snapshots, or focus issues.
create-cli-tool
Craft or refactor a CLI interface and behavior using the CLI Guidelines. Use when defining a new CLI spec or reshaping an existing command surface arguments, flags, subcommands, help text, outputs, errors, exit codes, prompts, config/env precedence, and safe/dry-run behavior.
execplan
When writing complex features or significant refactors or user ask explicitly, use an ExecPlan from design to implementation.
manipulate-xcodeproj
Manipulate Xcode projects and asset catalogs using XcodeProjectCLI (xcp) only. Use when editing .xcodeproj/.pbxproj files, adding/moving/renaming/deleting groups or files, changing target membership, reading or setting build settings, or managing xcassets assets.
notebooklm-superskill
Generate slide decks, audio podcasts, infographics, and video overviews from NotebookLM notebooks. Customizable by audience, format, language (80+), orientation, and visual themes. Use when asked to generate slides, create podcast, make infographic, video overview, or automate NotebookLM content creation.
common-pitfalls
Apply when debugging errors, reviewing code for issues, or encountering unexpected behavior. Contains known mistakes with ChurchTools API, Vue components, TypeScript, Safari cookies, and form handling.
development-workflow
Apply when setting up the development environment, running dev server, building, testing, or deploying the extension. Covers npm commands, CORS configuration, debugging, and deployment to ChurchTools.
Page 547 of 1645 · 82246 results