Claude Skills
The open format is called Agent Skills and works in Claude Code, Codex, Cursor and other agents — most people know it as Claude Skills.
Every Agent Skill we could find on GitHub, deduplicated by content. 79 437 files from 1 744 authors, of which 61 785 are unique — the rest is the same skill repackaged into someone else's repository. For each one: what it weighs in tokens, whether it ships runnable scripts, and which MCP servers it needs.
当设计或修改 REST API 响应结构、处理 API 返回值，或生成 Excel/CSV/PDF/对账文件等下游产物时触发。防止 API 设计缺陷导致的字段错位、类型歧义，以及生成产物时关键字段缺失但静默成功的问题。
>- 专业级 UI/UX 设计规范，需要高质量界面设计时手动触发或描述"设计感/专业UI"时自动触发。 覆盖视觉层次、配色体系、排版节奏、交互微动效、响应式适配等。 日常前端开发由 frontend-dev skill 覆盖。
当 API/任务可能执行超过 10 秒（批量数据处理、远程 API 批量调用、全表扫描、跨租户聚合）时触发。防止同步接口被网关 30s 超时切断、用户重复点击触发并发、状态缓存内存泄漏等问题。提供异步任务状态机标准模板。
Bash 脚本与系统命令规范。禁止行尾注释，强制使用 tee/heredoc 写入，提升命令执行的可维护性。
>- Python 开发规范。当用户操作 .py、pyproject.toml、requirements.txt、setup.py 文件， 或涉及 FastAPI、Django、Flask、pytest、asyncio 开发时触发。 包含 PEP 8 风格、类型注解、异常处理、测试规范、异步编程、性能优化等。
当编写新模块、设计接口、重构代码或代码审查时触发。提供经典模块化六原则检查清单（大小适中/调用深度/扇入扇出/边界清晰/作用域内聚/可预测性），适用于 PR/Review/新模块设计场景。
当重构涉及字段映射（dataIndex、枚举映射、类型转换）时触发。防止字段名推测错误，确保字段映射的正确性。
前端开发规范，包含 Vue 3 编码规范、UI 风格约束、TypeScript 规范等
Go 开发规范，包含命名约定、错误处理、并发编程、测试规范等。当操作 .go, go.mod, go.sum 文件时自动激活。
前端修改安全约束。当用户修改 Vue/React 组件、调整页面布局、创建覆盖层/蒙版组件时激活。提供数据绑定保护、布局一致性规范、覆盖层定位规范。
Java 开发规范，包含命名约定、异常处理、Spring Boot 最佳实践等
>- 当代码涉及多租户隔离（TenantContext、tenantId、租户拦截器/过滤器、X-Tenant-Code）时触发。 防止租户越权访问、数据串租户等安全问题。
运维安全详细规范。当涉及系统配置变更、核心服务重启、文件权限修改时激活。要求提供风险评估和回滚方案。
>- 当代码涉及支付回调、webhook 通知、notify_url、微信支付/支付宝回调处理时触发。 防止回调伪造、重放攻击、金额篡改等支付安全问题。
手绘卡通风信息图生成。用户描述内容后，Gemini 生成手绘涂鸦风格的信息图并保存到本地。手动触发。
Python 开发规范，包含 PEP 8 风格、类型注解、异步编程、性能优化等。当操作 .py, pyproject.toml, requirements.txt 文件时自动激活。
当代码涉及循环内查询、批量 ID 查询、IN 子句、BFS/递归遍历、嵌套 service 调用时触发。防止 N+1 查询、IN 子句过长、递归内存炸裂等性能陷阱。
商务专业风信息图生成。支持风格预设、配色、图标、布局、密度等参数化控制。手动触发。
当用户操作 Redis 相关代码（go-redis、Jedis、redis-py、ioredis）时触发。涵盖禁止 KEYS 命令、SCAN 替代、大 key 控制、Pipeline 批量、TTL 规范等安全与性能要点。
当用户要求重构代码（提取组件、合并重复逻辑、重命名变量、优化结构）时触发。提供重构安全检查清单，防止丢失原始上下文。
Rust 开发规范（聚焦 Axum + Tokio 后端栈），包含错误处理、异步陷阱、模块组织、测试规范等。当操作 .rs、Cargo.toml、Cargo.lock 文件时自动激活。
Review changed code for reuse, quality, and efficiency, then fix any issues found. Also scans project files for size limit violations.
>- 当代码出现 LocalDate.now() / LocalDateTime.now() / Instant.now() / new Date() / Date.now() / time.Now() 等"当前时间"调用，或涉及周/月/日起点、跨时区比较、数据库时间字段时触发。 防止生产环境时区漂移导致的统计错乱。
Converts books and documents (PDF, EPUB, DOCX, HTML, Markdown, plain text, RTF, MOBI/AZW with Calibre) into structured agent skills, extracting frameworks, mental models, principles, techniques, and anti-patterns. Use when the user wants to study a document through GitHub Copilot CLI, Amp, or Claude Code, apply an author's frameworks while working, or build a reusable knowledge base from a file.
> Use when troubleshooting an agent built with the Microsoft Agents SDK (Microsoft.Agents.Hosting.AspNetCore and related packages) in C# / .NET. startup, 401 or auth errors on incoming requests, the bot not responding to messages, appsettings.json configuration problems, Azure AD credential failures (AADSTS errors), port conflicts, or the agent not connecting in Teams or the Agents Playground. Use even if the user doesn't mention the SDK by name — trigger on symptoms like "my bot won't start", "getting 401s", or "bot isn't responding."
Use when migrating a Microsoft 365 Agents SDK agent that uses ActivityHandler or TeamsActivityHandler to AgentApplication. This is ONLY for DotNet projects that uses `Microsoft.Agents.*` packages. Triggered by Agents SDK bots that subclass ActivityHandler or TeamsActivityHandler that need to be modernized to AgentApplication routing.
> Use when any code imports Microsoft.Agents.Hosting.AspNetCore, Microsoft.Agents.Builder, or related Agents SDK packages, or when the user is building, configuring, or asking questions about a Microsoft 365 Agents SDK agent in C# / .NET. Trigger on questions about appsettings.json, connection configuration, AgentApplication patterns, OAuth sign-in flows, storage backends, cards, streaming, or local testing with the Agents Playground — even if no code exists yet and the user is planning or asking how to get started.
> Use when any code imports @microsoft/agents-hosting, @microsoft/agents-hosting-express, or related Agents SDK packages, or when the user is building, configuring, or asking questions about a Microsoft 365 Agents SDK agent in TypeScript. Trigger on questions about environment variables, connection configuration, AgentApplication patterns, OAuth sign-in flows, storage backends, cards, streaming, or local testing with the Agents Playground — even if no code exists yet and the user is planning or asking how to get started.
> Use when provisioning Azure resources for a Microsoft Agents SDK application — including creating an Azure Bot Service resource, setting up Entra app registrations, configuring identity credentials (UserManagedIdentity, FederatedCredentials, or ClientSecret), adding a Teams channel, running Bicep deployments for SSO, or setting up OAuth user sign-in connections via az CLI. Trigger even if the user doesn't say "provision" — use on requests like "set up my bot on Azure", "deploy my bot", "configure Azure for my bot", or "get Teams SSO working." Applies to all languages (dotnet, Node.js, Python).
> Use when troubleshooting an agent built with the Microsoft Agents SDK (@microsoft/agents-hosting and related packages). Trigger on any of these on incoming requests, the bot not responding to messages, .env configuration problems, Azure AD credential failures (AADSTS errors), port conflicts, or the agent not connecting in Teams or the Agents Playground. Use even if the user doesn't mention the SDK by name — trigger on symptoms like "my bot won't start", "getting 401s", or "bot isn't responding."
> Use when migrating a Bot Framework .NET SDK bot to Microsoft 365 Agents SDK. Microsoft.Bot.Builder.Integration.AspNet.Core that want to migrate to Agents SDK.
> Manages stacked PRs and splits multi-part work into reviewable branches with gh-stack. Use for stack creation, viewing, edits, push, submit, sync, rebase, merge, or checkout; when asked to split or isolate work for review; whenever a user mentions a stack, branch layers, dependent PRs, or gh stack; or when a stack is checked out.
| Use when the user asks about content in their OpenKB knowledge base — research topics, concepts compiled from their documents, cross-document synthesis — or mentions `openkb`, an `.openkb/` directory, or a `wiki/` tree generated by openkb. The user may invoke you from any working directory; the active KB resolves via `openkb status`. Do NOT use for arbitrary Markdown directories, Obsidian vaults, or documentation sites not built by openkb.
| Use to review a generated HTML deck or single-page artifact for visual quality and structural correctness. Especially good at catching CSS specificity bugs where slide-modifier classes (.divider, .center, .q, cause one slide to stack on top of every other. Also catches missing keyboard navigation, bullet-dump / wall-of-text failure modes, broken self-containment (external link/script/img). Patches the file in place; never changes the original content (slide text, numbers, named entities are the author's work, not yours).
| Use when the user asks the openkb chat to make a deck / slide presentation / PPT / slides / 演示稿 / 幻灯片 from their compiled KB content AND wants a dark, high-tech, neon / glow / glassmorphism look (赛博 / 科技风 / 暗色 / 霓虹 / 炫酷). Generates a polished single-file HTML deck in the Aurora Glass visual direction (near-black background, teal/sky/magenta/amber neon accents, glassmorphism panels, aurora gradient atmosphere) — opened in a browser, full-screened, shared. For the warm, printed, serif look use openkb-deck-editorial instead. Does NOT apply to generating skills (`openkb skill new`), research reports, or scrolling long-form documents.
| Use when the user asks the openkb chat to make a deck / slide presentation / PPT / slides / 演示稿 / 幻灯片 from their compiled KB content. Generates a polished single-file HTML deck in the Editorial Monocle visual direction (warm cream background, serif type, brick-red accent) — designed to be opened in a browser, full-screened, and shared. Does NOT apply to generating skills (that's `openkb skill new`), long-form research reports, or interactive prototypes.
Answer in the ADHD-friendly Attention-kind style for the rest of this chat.
Answer in the Rundown briefing style (TL;DR + checklist + numbered choices).
Compress a document, thread, transcript, or pasted text into a scannable TL;DR.
Answer in the terse, zero-warmth Spartan style for the rest of this chat.
Expert code review of current git changes with a senior engineer lens. Detects SOLID violations, security risks, and proposes actionable improvements.
Use when user requests diagrams, flowcharts, architecture charts, or visualizations. Also use proactively when explaining systems with 3+ components, complex data flows, or relationships that benefit from visual representation. Generates .excalidraw files and exports to PNG/SVG via Kroki API or locally using excalidraw-brute-export-cli.
Best practices for Remotion - Video creation in React
Use when the user gives a topic and wants an automated video podcast created, or asks to learn visual design patterns from a reference video/image. Produces 4K video via research → script → TTS → Remotion → MP4 + BGM.
小遥搜索 MCP 工具 - 本地文件智能搜索（语义/全文/图像/语音/混合搜索）
> Turn ONE topic into a finished Vox-style paper-collage explainer / ad video, end to end on the Atlas Cloud API + local ffmpeg — script, collage keyframes, motion, voice-over, music, captions, all automated. Use this whenever the user wants a "Vox style" video, a paper/torn-paper collage animation, a "motion collage", a narrated explainer or short ad built from AI-generated collage posters, a scrapbook-style tribute, or wants to turn a topic / product / person into a punchy narrated collage video — even if they don't say the word "Vox". Also use when reproducing Stav Zilber / rom1trs / Higgsfield-style collage ad workflows. photo of a person/product anchored into the collage (C-roll mode). explainer", "make a collage ad", "turn this topic into a collage video", "turn my photo/this product shot into a collage video".
Use ctx to search local coding-agent history before acting. Use when prior agent sessions may contain relevant insights, decisions, attempts, or transcript context.
Find Japanese NLP GitHub repositories and Hugging Face models/datasets that are NOT yet in awesome-japanese-nlp-resources. Suggests candidates to add for a given topic using WebSearch + WebFetch, then outputs contribution-ready markdown.
Investigate current challenges, limitations, and proposed solutions in Japanese NLP for a topic. Surveys the existing awesome-japanese-nlp-resources dataset to see what already exists, then web-researches known problems and ongoing efforts to produce a digestible issue report.
Analyze current trends in Japanese NLP for a topic. Surveys the existing awesome-japanese-nlp-resources dataset and augments it with up-to-the-minute web research to produce a digestible trend report.
Search all Japanese NLP resources (libraries, models, datasets, tutorials, dictionaries, Hugging Face). Accepts keywords or natural language questions in any language.
Given a Japanese NLP GitHub repo or Hugging Face model/dataset (URL / owner/repo / tool name), find repositories or models/datasets that do the same or related processing. Mines the bundled dataset for content-similar items, then expands via web research across both GitHub and Hugging Face, then merges and re-ranks.
Generate a lean AGENTS.md hub file for a repository or area, including recommended topics for detail files.
Generate a concise .github/copilot-instructions.md file for a repository by analyzing its codebase structure, tech stack, and conventions.
Generate a concise .instructions.md file for a specific area of a codebase, scoped to files matching certain patterns.
Generate a deep-dive instruction file about a specific topic for a repository or area.
Modify, build, test, debug, and contribute to NVIDIA cuOpt (C++/CUDA, Python, server, CI). Use for solver internals, PRs, DCO, and code conventions.
Trace, complete, and interpret the Pareto frontier across competing objectives using repeated single-objective cuOpt solves (weighted-sum and ε-constraint).
Install cuOpt for Python, C, or server via pip, conda, or Docker; verify the install. For building cuOpt from source, see cuopt-developer.
cuOpt REST server — start server, endpoints, Python/curl client examples. Use when the user is deploying or calling the REST API.
Claude Skills — questions
Answers built from the skills we actually parsed.