Coding & Development

benchtop

github.com

Take a microcontroller or 3D-printing project from a loose idea to a package someone can build - concept, parts and live prices, a checked circuit, a carrier PCB or an illustrated flat-pack-style soldering guide drawn to scale, a 3D-printable enclosure, firmware scaffold, and a designed landing page with every file. Use this whenever the user wants to make a physical electronics thing (ESP32, ESP8266, Arduino, Raspberry Pi Pico, XIAO, sensors, displays, LEDs, buttons, servos, batteries), asks what parts to buy, how to wire or solder something, whether to order a PCB, wants a case or enclosure for electronics, wants a wiring diagram, BOM, assembly instructions or a build guide, or wants to present or share a hardware project - even if they only say "I want to build a thing that..." and never mention a microcontroller.

$ npx skills add filipjaj/benchtop
Skill anzeigen→

swiftui-expert-skill

github.com

Use when writing, reviewing, or refactoring SwiftUI code for iOS or macOS, including state and `@Observable` data flow, view composition, resizable layouts, safe areas, display scale, performance, lists, environment, localization, animation, Liquid Glass, and API migration. Also use for iPhone Duo, foldable, or large-display layouts (`NavigationSplitView` on large displays, iPhone tab sidebar, two-column reflow, foldable grids, `ArrangementView`, `ReservedRegion`), hinge effects, vertical bars, `@State` initialization or synthesized-property diagnostics, `@ContentBuilder` ambiguity, `reorderable` drag/drop, custom `AsyncImage` `URLSession`, swipe actions outside List, item-bound `alert`/`confirmationDialog`, `ToolbarOverflowMenu`, `AnimatableValues`, Document APIs (`Document`/`DocumentReader`), and Instruments `.trace` capture or analysis.

$ npx skills add erumkatzwan/agentic-product-starter
Skill anzeigen→

threejs-gltf-loading

github.com

Loads glTF/GLB models into three.js r187 and plays their animations: GLTFLoader, DRACO, Meshopt and KTX2 decoders, AnimationMixer cross-fades and SkeletonUtils clones. Use when a model is invisible or tiny, decoder files 404, animations never play, or clones share one skeleton. Not for scene setup (`threejs-scene-setup`) or lighting (`threejs-materials-lighting`).

$ npx skills add mext-ai/elearning-agent-skills
Skill anzeigen→

olakunlevpn-interaction-map

github.com

Use BEFORE writing code for any feature that has a user-facing flow and a server leg -- "build X", "add X feature", "design the X flow", "how should this work", "we need a page that does Y". Designs the interaction flow as three exhaustive lists (user steps with every state decided, the server chain, and the seam contract between them) and writes them to a spec file before either side is built. Decides the loading, empty, error and forbidden states rather than discovering later that nobody designed them, and fixes the exact wire shape before the client and server can drift. Standalone -- depends on no other skill. Gate is countable and passes only at 1.0. Do NOT use when the flow already exists in code (that is an audit, not a design), or for pure styling and copy work.

$ npx skills add olakunlevpn/olakunlevpn-interaction-map
Skill anzeigen→

frontend-verification

github.com

改完网页/前端界面(CSS、布局、组件、动画、响应式、深色模式、打印)之后怎样验证并拿出证据:先用内置浏览器预览,覆盖不到时用无头 Chrome 截图或读 DOM(预览面板被隐藏、批量截多个视口、要先进入某个状态、要拿 DOM 结果),附可直接运行的 shot.sh / shot.mjs 脚本。还收录无头 Chrome 的坑(requestAnimationFrame 动画停在第一帧、375 宽被撑成 500、深色模式跟随系统)、字体不显示或像宋体的排查、整页截图里滚动动画区块空白、vitest 排除 .claude worktree,以及桌面/手机/英文/繁體/深浅色/打印的核对清单。只要改了界面、要截图、要查响应式或深色模式、动画不动,或者准备说“界面改好了”,就用这个 skill,哪怕用户只说“看看效果”。Use after any UI/CSS/layout/animation change to verify it, take screenshots at several viewports, check dark mode / mobile / print, or debug animations frozen in headless Chrome.

$ npx skills add RedCarpG/redcarpg-skills
Skill anzeigen→

vrm-model-conversion

github.com

把带骨架的 3D 人物模型(FBX、PMX/MMD、XPS、glTF/GLB、.blend;来自 Mixamo、MMD、Unity/VRChat、Daz Genesis、Auto-Rig Pro、Rigify、3ds Max Biped、游戏提取)在 Blender 后台转成 VRM 1.0,给 three-vrm 网页、桌宠、VTuber 软件、VRChat 外的应用用。管:解压和挑哪份源文件、各格式怎么导入、骨头名自动对应到 VRM 标准骨骼、A-pose 转 T-pose、表情(已有形态键直接绑;只有脸部骨头的烘成形态键)、头发和衣服的弹簧骨骼(没有物理骨的就加骨链、按位置重新分权重)、碰撞体、MToon 材质和发片透明、贴图缩小、VRM meta 和许可,以及一份检查清单。附可复用的 Blender 脚本 vrm_kit.py(参数化,不写死文件名)、模板 build_template.py、不用 Blender 的检查脚本 inspect_vrm.py。只要用户想「把这个模型转成 VRM」「这个 FBX/PMX 能不能给 three-vrm 用」「给模型加表情、加头发物理、加弹簧骨」「VRM 的表情不动、头发不飘、裙子穿模、碰撞体没用」「下载了个模型想当桌宠/虚拟形象」,就用这个 skill,即使没说 VRM。Also use for English requests to convert FBX/PMX/glTF/blend avatars to VRM, set up VRM expressions, spring bones or colliders, or debug a VRM export from Blender.

$ npx skills add RedCarpG/redcarpg-skills
Skill anzeigen→

react-native-skia

github.com

Build, debug, and investigate React Native Skia canvases, paths, shaders, graphics animation, and rendering performance. Use for code using @shopify/react-native-skia or react-native-skia, including Skia integration with Expo, Reanimated, Worklets, and Gesture Handler. Do not use for unrelated React Native layout, navigation, or backend tasks.

$ npx skills add gokhancaliskan/react-native-skia-skills
Skill anzeigen→

undo-first

github.com

Reversible-action infrastructure for AI agents. Use before any side-effecting action (writing/deleting/moving files, config changes, git pushes, sending messages, deploys, API calls) when the user asks "can we roll this back?", "make this safe to try", or when executing risky multi-step work. Provides hash-verified snapshots, a risk gate that demands an undo or compensation plan before acting, byte-identical restore, and an audit journal. Philosophy - no undo path, no action.

$ npx skills add thiennhant95/agent-skills
Skill anzeigen→

skill-firewall

github.com

Security scanner for AI agent skills (SKILL.md packs). Use before installing any skill from ClawHub, GitHub or external sources, when auditing a skill directory for malicious patterns (prompt injection, credential theft, exfiltration, destructive commands, obfuscated payloads, persistence), or when asked "is this skill safe?". Produces a SAFE/CAUTION/SUSPICIOUS/MALICIOUS verdict with evidence and CI-friendly exit codes.

$ npx skills add thiennhant95/agent-skills
Skill anzeigen→

scrum-spec

github.com

HUMAN-ONLY COMMAND. Drafts a Scrum team's spec, one spec.md per story - stories from a flow, acceptance criteria and open questions grounded in the code, test steps for the tester. Finds facts in the code, asks people only for decisions, records who decided. Type /scrum-spec, e.g. /scrum-spec ac rb-12-book-room.

$ npx skills add thitinan147/scrum-spec
Skill anzeigen→

project-init

github.com

Initialize or upgrade a software project to a universal documentation standard. Runs git init, generates PRD.md / TRD.md / EXECUTED.md at the root, builds a four-audience docs/ knowledge base (ai, technical, layman, non-technical), writes an AGENTS.md conventions file, and in full mode documents every folder and file with header comments carrying doc references, ISO-8601 timestamps, and author. Use when creating, initializing, or scaffolding a new project or repository, or when asked to document an existing codebase to this standard.

$ npx skills add mridun-gupta/project-init.skills
Skill anzeigen→

advisor

github.com

Pair yourself (the executor) with a stronger reviewer model that you consult at a few key moments of a coding task - before committing to an approach, when stuck, and before declaring done. Use this skill on any coding or agentic task longer than a few steps, on design, architecture and risk questions, when errors keep recurring or an approach is not converging, and whenever the user mentions an advisor, a second opinion, a stronger model, or a plan review. Works in any coding agent that can start a subagent or a second model session.

$ npx skills add RogerioTrani/coding-agent-advisor-skill
Skill anzeigen→

system-design

github.com

Plans and maintains the project's system design as a live map (n8n-style nodes and connections, limits, architecture recommendations and dependency CVEs) on a local page. Use it to map an existing system and to design the backend of a new app from scratch. Use it when the user asks for system design, an architecture map, a connection diagram, capacity limits, scaling or dependency vulnerabilities, when planning the backend of a new app, and before any task that changes the flow (new service, database, queue, cache, external API, authentication, deploy) to propose it and get approval on the map before implementing.

$ npx skills add CarlosAllberto/system-design-skill
Skill anzeigen→

codebase-explainer

github.com

Analyze a codebase and write a Markdown guide (CODEBASE_GUIDE.md) that explains what every file does, why the app needs it, and how the pieces connect, with Mermaid diagrams. Use when the user wants to understand, learn, map, document or onboard onto a repo, or asks "what does each file do" or "how does this app work".

$ npx skills add DarenVong/Codebase-explainer
Skill anzeigen→

java-swing

github.com

Build, design, fix, review and test Java Swing desktop apps so they look modern and stay fast. Use when code imports javax.swing or java.awt, or the task mentions Swing, JFrame, JTable, JList, JTree, the EDT, FlatLaf, look and feel, a dated or ugly UI, a frozen or flickering window, HiDPI/display scaling, dark mode, custom painting, big data in tables or lists, UI/visual tests, xvfb, or packaging a Java desktop app with jpackage.

$ npx skills add bdarwin/java-swing-skill
Skill anzeigen→

lead-agent-workflow

github.com

Lead coding work through worker agents from a request to merged PRs: build a task graph, dispatch coders and reviewers, run the review loop, gate on CI, merge when allowed, and keep the session alive and cheap. Works with any orchestration tool (Paseo, T3 Code, Claude Code, others). Use only when the user explicitly asks you to orchestrate, coordinate or dispatch other agents to do the work, e.g. "orchestrate agents to fix these issues", "dispatch coders and reviewers", "use Paseo to run this". Do not use when the user asks you to do the work yourself, or for a single helper subagent you start on your own initiative.

$ npx skills add duykhanh01/agent-skills
Skill anzeigen→

write-pr

github.com

Write or update a pull request title and description. Use when opening a PR, filling in a PR template, or revising a PR description after the change has moved on.

$ npx skills add duykhanh01/agent-skills
Skill anzeigen→

engineering-team

github.com

Select and run a balanced engineering team for a software brief. Use when planning, designing, or making implementation decisions on work with a real design choice, from small features to multi-area or high-risk changes. Not for one-file fixes, clear bugs or small edits.

$ npx skills add SamNewhouse/engineering-team
Skill anzeigen→

feature-sliced-nx

github.com

Feature-Sliced Design for an Nx workspace. Each slice is a React library under libs/<domain>/<layer>/<slice>, generated with fsd-slice (wraps @nx/react:library), tagged fsd:layer and fsd:domain, with pages-first placement and widgets discouraged. Use when the repo has nx.json and the task creates, moves, splits, or imports a slice, decides where code belongs, refactors an existing app onto these libraries, adds Storybook, or wires the harness. Do not use the agnostic feature-sliced-design src/ layout in an Nx workspace.

$ npx skills add effinrich/agent-skills-fsd
Skill anzeigen→

feature-sliced-design

github.com

Agnostic Feature-Sliced Design for a single React app (src/app, pages, features, entities, shared; widgets discouraged), pages-first placement, slice public APIs, downward imports, and the gate that enforces them. Use when the repo has no nx.json and the task creates a slice, decides where code belongs, splits a component, or moves an existing React app onto these layers. For an Nx workspace, use the feature-sliced-nx skill instead.

$ npx skills add effinrich/agent-skills-fsd
Skill anzeigen→

clean-project

github.com

How to build and change software cleanly, with the person — understand an existing codebase before touching it, ask the right questions before a new project, agree what the agent may decide alone, raise overlooked risks, choose the right technologies before building, start from a fitting scaffold with automated format, lint and CI, keep code simple with no clutter, dead code or duplicate paths, keep Git clean, and put each change in the module that owns it. Use whenever writing or changing code in a repository — a new app, a stack choice, a feature, a bug fix, a refactor, a cleanup, a project health check, or setting up linters or CI — even if the user never says "clean". Requests in any language count, e.g. «سو لي مشروع»، «ابي اضيف ميزة»، «صلّح الخطأ»، «نظّف المشروع»، «افحص المشروع». Not needed for one-line edits, renames, explaining code, or non-code writing.

$ npx skills add 0xasem/agent-skills
Skill anzeigen→

explain-stack

github.com

Explain in detail WHY a project uses each technology, for beginners and vibe coders. Scans the current project and, for every framework, library, database and tool, explains what it is (with an analogy), why and how it is used here (file:line, key config), version notes, alternatives, gotchas and what would break without it. Adds a project map, how to run it, a request traced from click to database, the data model, configuration, findings, a glossary, a quiz and STACK.md. "/explain-stack redis" deep-dives one technology; "quick" gives a short overview; "code [path]" walks through key code and algorithms line by line; "changes [commit|N|branch]" explains what just changed (e.g. after an AI agent edited files), with risks and how to test it. Use when the user asks "why does this project use X", "explain my stack", "what is all this in package.json", "what did the AI just change", "explain this code", "how does this algorithm work", "objasni mi tehnologije", "zašto koristimo X",

$ npx skills add Aleksa5000/explain-stack
Skill anzeigen→

chinese-tts

github.com

让程序把中文(夹英文、数字、链接、代码)念出来时的文本规范化(TN)和语音合成工程:先转念法再切句、流式输出时怎么扣住数字边界、年份/电话/验证码/单位/温度/百分比/范围/分数/网络用语怎么念、念法替换表、念错了怎么查;本机 TTS 引擎(Kokoro、Matcha、CosyVoice、Qwen3-TTS、sherpa-onnx、vLLM、WSL2)的选型、首句延迟、取消、缓存、后备音源、显存;语音输入和插话。附 139 条验收过的「原文 → 念法」对照用例和一个检查脚本。只要用户在做朗读、语音助手、AI 角色说话、有声读物、播报、TTS 前处理,或者抱怨「数字念错了」「年份念成两千零二十六」「电话号码念成几亿」「链接被逐字念」「英文被跳过」「流式念出来和整句不一样」「取消以后下一句很慢」,就用这个 skill,即使用户没说「TN」或「文本规范化」。Also use for English requests about Chinese text normalization for TTS, number/date reading in Mandarin speech synthesis, or streaming TTS latency.

$ npx skills add RedCarpG/redcarpg-skills
Skill anzeigen→

secure-webapp

github.com

Use for OWASP-grounded security guidance when building, editing, or reviewing web applications. Trigger proactively for auth, sessions, JWT/OAuth/OIDC, user input, DB queries, file uploads, API endpoints, cookies/CORS/CSP/security headers, secrets, redirects, external URL fetches, logging/errors, dependencies, threat modeling, hardening, audits, LLM/AI features such as prompt handling, agent tools, and MCP servers, or vulnerability classes such as XSS, SQLi, IDOR, CSRF, SSRF, open redirect, prototype pollution, deserialization, prompt injection, and supply chain risk.

$ npx skills add gabrielereylane/lovable-skills
Skill anzeigen→