claude_code_template

构建教材级别的黑白矢量插图（几何图、受力分析图、光路图、电路图、函数图像、坐标系、物理/数学示意图等），以自包含 HTML 文件呈现——所有线条用 SVG 绘制，所有公式、符号、运算符、物理/数学符号用 LaTeX 经 MathJax 渲染为矢量，支持一键导出高清 PNG。当用户说"画个教材图/教材插图/教学示意图""画受力分析图/光路图/电路图/函数图/几何图""做一张黑白矢量图/示意图""把这道题/这个原理配张图""画个坐标系/数轴"，或任何为教材、讲义、试卷、课件绘制精确黑白教学插图的需求时，必须使用本技能，即使用户没明说"教材"二字。注意：本技能与 make-infographic 不同——make-infographic 做彩色信息图/海报（给家长客户看，版式驱动），本技能做黑白精确教学插图（给师生用，准确性驱动），不要混用。

调用 ZenMux 中转平台的 openai/gpt-image-2 模型生成 AI 图片（文生图）。当用户要求"生成图片/画一张图/AI画图/做张图/文生图/出图"，或需要把文字描述、提示词变成图片时，必须使用本技能，即使用户没明说"用 AI 生成"。流程为：设计提示词（确定主题/内容/主体元素/尺寸/清晰度，缺失项先用 AskUserQuestion 询问）→ 把完整提示词与参数展示给用户确认 → 确认后调用 scripts/generate_image.py 生成，保存到 AI_output 且文件名与内容相符 → 用视觉模型检查主体是否清晰无遮挡、文字是否完整 → 有问题告知用户，由用户决定是否重新生成。

把信息、数据、规划、流程、知识等内容转化为家长/客户能看懂的可视化图像（信息图、流程图、脑图/思维导图、关系图、知识图谱、海报），以自包含 HTML 文件呈现，所有公式用 LaTeX 编写并实时渲染，支持一键导出高清 PNG。当用户说"做成信息图/可视化/图片/海报""画个流程图/脑图/思维导图/关系图""把这份内容做成图给家长/客户看"，或任何把内容可视化成图的意图时，必须使用本技能，即使用户没明说"信息图"二字。

End-to-end backend engineering workflow for building backend services, creating, modifying, optimizing, debugging, testing, and documenting server endpoints, detecting backend risks, refactoring or rewriting backend code, and verifying server behavior. Use when the user asks to build a backend, create a server, implement backend features, add or adjust REST APIs, GraphQL resolvers, RPC handlers, controllers, routes, request validation, response schemas, authentication, authorization, pagination, filtering, sorting, OpenAPI/Swagger specs, backend integration tests, performance/error handling, database logic, or backend quality hardening. Do NOT use for frontend-only work, deployment-only tasks with no backend code change, database-only analysis with no service change, or high-level architecture discussion with no implementation.

Build, debug, and optimize Claude API / Anthropic SDK apps. Apps built with this skill should include prompt caching. Also handles migrating existing Claude API code between Claude model versions (4.5 → 4.6, 4.6 → 4.7, retired-model replacements). TRIGGER when: code imports `anthropic`/`@anthropic-ai/sdk`; user asks for the Claude API, Anthropic SDK, or Managed Agents; user adds/modifies/tunes a Claude feature (caching, thinking, compaction, tool use, batch, files, citations, memory) or model (Opus/Sonnet/Haiku) in a file; questions about prompt caching / cache hit rate in an Anthropic SDK project. SKIP: file imports `openai`/other-provider SDK, filename like `*-openai.py`/`*-generic.py`, provider-neutral code, general programming/ML.

End-to-end database engineering workflow for designing relational data models, writing schemas and migrations, reviewing constraints, creating indexes, optimizing queries, and hardening persistence behavior. Use when the user asks to design a database, create or change tables, choose fields/types/keys, model entity relationships, write SQL/ORM migrations, review database schema quality, improve slow queries, add indexes, plan transactions, handle multi-tenant data, or evaluate data integrity/security risks. Do NOT use for frontend-only work, one-off spreadsheet analysis, or high-level product brainstorming with no persisted data model.

End-to-end frontend development workflow for building or modifying web apps, pages, components, dashboards, HTML/CSS/JS interfaces, React/Vue/Svelte projects, or UI artifacts. Use when the user asks to create, redesign, implement, fix, or polish frontend work and expects the model to plan first, orchestrate the implementation, develop the code, then verify it in a browser. Trigger for requests about frontend development, page development, component development, dashboards, web applications, UI implementation, responsive styling, or visual polish. Do NOT use when the user only asks for backend/API/database work, a design critique with no code changes, or a high-level explanation with no implementation.

大规模变更任务规划技能。当主程序需要对项目进行较大变动时自动激活，包括：大幅度修改功能模块代码、添加新功能或新模块、大规模优化或重构、重写代码等。不适用于单文件小修小补或简单的 bug 修复。