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. 79 566 files from 1 758 authors, of which 61 913 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.
Get started building on Shopify. Use when a developer asks to build an app, build a theme, create a dev store, set up a partner account, scaffold a project, or get started developing for Shopify. NOT for merchants managing stores.
Build custom functionality that merchants can install at defined points in the checkout flow, including product information, shipping, payment, order summary, and Shop Pay. Checkout UI Extensions also supports scaffolding new checkout extensions using Shopify CLI commands.
Build custom functionality that merchants can install at defined points on the Order index, Order status, and Profile pages in customer accounts. Customer Account UI Extensions also supports scaffolding new customer account extensions using Shopify CLI commands.
Choose when the user needs **Shopify CLI** to run or fix something now: validate app or extension config on disk (`shopify.app.toml`, `shopify.app.<name>.toml`, `shopify.extension.toml`); run or troubleshoot store workflows (`shopify store auth`, `shopify store execute`); or perform explicit store-scoped reads/writes on a named store domain (for example, show/list/find the first 10 products on my store at `foo.myshopify.com`, or inventory and product changes by handle, SKU, or location name). Emphasize **commands and operational steps**, not only authoring GraphQL. Skip for API-only understanding or codegen with no CLI execution, and skip for brand-new merchant asks to start a Shopify store or try Shopify before they have an account. Examples: validate configuration before deploy; run an existing query via CLI; show the first 10 products on `foo.myshopify.com`; missing `shopify store execute`.
Build retail point-of-sale applications using Shopify's POS UI components. These components provide a consistent and familiar interface for POS applications. POS UI Extensions also supports scaffolding new POS extensions using Shopify CLI commands. Keywords: POS, Retail, smart grid
Use for custom storefronts requiring direct GraphQL queries/mutations for data fetching and cart operations. Choose this when you need full control over data fetching and rendering your own UI. NOT for Web Components - if the prompt mentions HTML tags like <shopify-store>, <shopify-cart>, use storefront-web-components instead.
Answer a merchant's **analytics and reporting** questions with **ShopifyQL** — Shopify's query language for aggregated store metrics that the Admin GraphQL API cannot compute. Choose this (not `admin`) whenever the ask is for **numbers, totals, trends, or breakdowns** rather than fetching or mutating individual records: including but not limited to total/gross/net sales and revenue, order counts, average order value, refunds, quantity sold, sessions, conversion rate, and traffic — sliced by product, channel, region, or customer, trended over time, or compared period-over-period. Examples: \"total sales last 7 days\", \"orders by sales channel this month\", \"top products by revenue\", \"conversion rate this week\", \"sales this year vs last year\". This topic covers writing the ShopifyQL query; if the merchant wants to run it against their store, execution is handed off to `use-shopify-cli`. Not for general Admin GraphQL record operations — fetching or mutating individual resources (use `admin`).
Rigor Reproduce compatible skill slug for README-first deep learning repository reproduction. Use when the user wants an end-to-end, minimal-trustworthy flow that reads the repository first, selects the smallest documented inference or evaluation target, coordinates intake, setup, trusted execution, optional trusted training, optional repository analysis, and optional paper-gap resolution, enforces conservative patch rules, records evidence assumptions deviations and human decision points, and writes the standardized `repro_outputs/` bundle. Do not use for paper summary, generic environment setup, isolated repo scanning, standalone command execution, silent protocol changes, score chasing, or broad research assistance outside repository-grounded reproduction.
Rigor Explore compatible skill slug for meaningful and potentially novel deep learning research candidates. Use when the researcher has chosen the task family, dataset, benchmark, evaluation method, provided SOTA references, and wants candidate-only exploration on top of `current_research` with auditable repo understanding, idea gating, fair comparison, and governed experiments written to `explore_outputs/`. Do not use for README-first trusted reproduction, open-ended direction finding, narrow code-only or run-only exploration, passive repo analysis, verified novelty claims, or implicit experimentation.
Rigor Run skill for README-first deep learning repo reproduction. Use when the task is specifically to capture or normalize evidence from the selected smoke test or documented inference or evaluation command and write standardized `repro_outputs/` files, including patch notes when repository files changed. Do not use for training execution, initial repo intake, generic environment setup, paper lookup, target selection, hidden scientific-meaning changes, or end-to-end orchestration by itself.
Rigor Setup skill for README-first deep learning repo reproduction. Use when the task is specifically to prepare a conservative conda-first environment, checkpoint and dataset path assumptions, cache location hints, and setup notes before any run on a README-documented repository. Do not use for repo scanning, full orchestration, paper interpretation, final run reporting, or generic environment setup that is not tied to a specific reproduction target.
Rigor Paper Context helper for README-first deep learning repo reproduction. Use only when the README and repository files leave a narrow reproduction-critical gap and the task is to resolve a specific paper detail such as dataset split, preprocessing, evaluation protocol, checkpoint mapping, or runtime assumption from primary paper sources while recording conflicts. Do not use for general paper summary, repo scanning, environment setup, command execution, title-only paper lookup, or replacing README guidance by default.
Rigor Analyze / Rigor Audit read-only skill for deep learning research repositories. Use when the user wants to read and understand a repository, inspect model structure and training or inference entrypoints, review configs and insertion points, or flag suspicious implementation patterns without modifying code or running heavy jobs. Do not use for active command execution, broad refactoring, speculative code adaptation, or automatic bug fixing.
Rigor Improve implementation leaf skill for auditable candidate implementation in deep learning research repositories. Use when the researcher explicitly authorizes exploratory work on an isolated branch or worktree to transplant modules, adapt a backbone, add LoRA or adapter layers, replace a head, or stitch together meaningful low-risk migration ideas with rollback-aware records in `explore_outputs/`. Do not use for end-to-end exploration orchestration on top of `current_research`, trusted baseline reproduction, conservative debugging, environment setup, verified contribution claims, or default repository analysis.
Rigor Improve / Rigor Explore run leaf skill for bounded exploratory evidence in deep learning research repositories. Use when the researcher explicitly authorizes exploratory runs such as small-subset validation, short-cycle guess-and-check, batch sweeps, idle-GPU search, or quick transfer-learning trials, with fair-comparison caveats and no-overclaim summaries in `explore_outputs/`. Do not use for end-to-end exploration orchestration on top of `current_research`, trusted baseline execution, conservative training verification, default routing, verified SOTA claims, or implicit experimentation.
Rigor Intake helper for README-first deep learning repo reproduction. Use when the task is specifically to scan a repository, read the README and common project files, extract documented commands, classify inference, evaluation, and training candidates, and return the smallest trustworthy reproduction plan to the main orchestrator. Do not use for environment setup, asset download, command execution, final reporting, paper lookup, or end-to-end orchestration.
Rigor Debug / Rigor Audit skill for deep learning research work. Use when the user pastes a traceback, terminal error, CUDA OOM, checkpoint load failure, shape mismatch, NaN loss symptom, or training failure and wants conservative diagnosis before any patching, with debug fixes clearly separated from research contributions. Do not use for broad refactoring, speculative adaptation, automatic exploratory patching, or general repository familiarization.
Rigor Train skill for deep learning research repositories. Use when a documented or selected training command should be run conservatively for startup verification, short-run verification, full kickoff, or resume, with command, config, seed, log, checkpoint, status, and metric evidence written to standardized `train_outputs/`. Do not use for environment setup, exploratory sweeps, speculative idea implementation, or end-to-end orchestration.
规划、审计和优化合规的抖音新号起号与涨粉系统，覆盖账号定位、观看理由、标签校准、搜索流量预埋、3秒钩子、人格化表达、评论互动、私域冷启动、合集运营、9条视频小样本实验、数据复盘和30天执行计划。用于用户询问抖音起号、抖音涨粉、新号冷启动、账号定位、低播放量、对标账号、短视频选题、完播率、评论互动、合集策略、私域启动或抖音内容复盘时。
规划、审计和优化合规的微信公众号起号系统，覆盖账号定位、主页框架、对标拆解、选题库、文章简报、发布节奏、流量主实验和周复盘。用于用户询问公众号起号、新号冷启动、流量主、公众号定位、对标账号、爆款标题、30天起号计划或公众号内容策略时。
当用户需要创建、诊断或优化小红书账号起号，包括起号、做小红书、涨粉、引流、账号定位、个人 IP、选题库、对标分析、笔记简报、内容日历、主页诊断、转化路径或 30/60/90 天起号计划。
中文 X/Twitter 起号与冷启动专家。用于帮助普通人从 0 到早期正反馈搭建账号定位、内容主题、回复区曝光、主贴/Thread 转化、7 天执行计划、周复盘表和增长诊断。触发场景包括：用户想做 X 账号冷启动、推特起号、500 粉以内增长、个人 IP 定位、内容矩阵、互动回复策略、将真实工作流转成内容、复盘 X 数据，或要求基于账号/选题/截图制定可执行起号方案。
规划、审计和优化合规的微信视频号从 0 到 1 起号系统，覆盖账号定位、人设打造、选题策划、视频脚本、私域承接、9条视频小样本实验、数据复盘和30天执行计划。用于用户询问视频号起号、视频号涨粉、新号冷启动、账号定位、低播放量、对标账号、视频号脚本、完播率、评论互动或视频号内容复盘时。
XDP-for-Windows test execution and debugging workflow. Use when running tests (functional, spinxsk, pktfuzz, ringperf, rxfilter, xskmaprx, xskfwdkm), debugging bugchecks, attaching kd, recovering test machines after crashes, or invoking any tools/*.ps1 script that runs on hardware. Covers PowerShell remoting via -ComputerName, kd remote pipe debugging, and the inject/build/bugcheck/.reboot/check-drivers recovery loop.
Apply a visual direction — an archetype (high-end agency, editorial minimal, brutalist, soft-SaaS, dark-tech) or one of 138 named design systems (apple, linear-app, stripe, vercel, notion, material, shadcn, spotify, tesla…) — by resolving it into the token system. Use when the user wants a specific look/vibe/brand feel, or asks to make a design feel premium/expensive/non-generic.
Generate, extend, or audit design tokens in DTCG format with the 3-tier architecture (primitive → semantic → component). Use when the user wants a color palette, type scale, spacing/shadow/radius/motion tokens, multi-brand theming, or wants to validate token files. Covers colors, typography, spacing, shadows, borders, breakpoints, motion, gradients, opacity, blur, sizing, states, theming.
Generate a complete, accessible brand design system from a brief — primitive → semantic → component DTCG tokens (color, type, spacing, radius, shadow, motion), light + dark, plus a single theme.css — verified for WCAG. Use when the user wants a from-scratch brand/design foundation, a new palette + type system, or a themeable token kit for a product.
Design a UI component spec to the house quality bar — anatomy, variants, sizes, the 8 states, token mapping, and accessibility. Use when the user wants to design or document a component (button, input, tabs, toast, combobox, date picker, modal, etc.) at the spec level before or alongside code. For generating framework code, use design-code.
Audit a UI or design against WCAG 2.2 AA/AAA and ARIA patterns, returning criterion-referenced findings with severity and specific fixes. Use when the user wants an accessibility check, contrast verification, keyboard/screen-reader review, or wants to confirm a component meets POUR.
Generate production-ready, accessible, token-driven component code for ANY framework — React+Tailwind, Next.js, SwiftUI, Vue, Svelte, Angular, Solid, Web Components/Lit, React Native, Flutter, Jetpack Compose, vanilla CSS, or CSS-in-JS. Use when the user wants working UI code for a component or screen in a specific stack.
Set up or run design QA gates — token + hardcoded-value lint, automated a11y (axe), contrast, visual regression across variants/states/themes/RTL, and the manual a11y checklist. Use when the user wants CI quality gates, to prevent design regressions, or to QA a component/screen before shipping.
Review or audit a design/UI across 6 weighted dimensions with Nielsen's 10 heuristics and a prioritized findings table. Use when the user wants a design critique, quality score, heuristic evaluation, or audit of an existing screen, page, or product before/after build.
Keep Figma and code in sync — map the 3-tier DTCG tokens to Figma Variables (collections + modes), sync in either direction, use the Figma MCP when connected, and verify component parity (variants/states). Use when the user wants to push tokens/components to Figma, pull a design into code, set up token↔Variable sync, or check design-code drift.
Govern how the design system evolves — SemVer for tokens/components, the contribution workflow, deprecation policy, and change communication. Use when the user wants to add/promote/deprecate a component or token, decide a version bump, set up a contribution process, or keep the system from fragmenting.
Optimize UI performance against Core Web Vitals — LCP, INP, CLS — with loading/code-split strategy, layout-shift prevention, and animation performance rules. Use when the user wants to improve speed, fix jank or layout shift, hit Web Vitals budgets, or make a UI feel fast on low-end devices.
Move an idea up the fidelity ladder (content-first → wireframe → low-fi → high-fi → code) with a validation plan at each level, plus user-journey mapping and usability-testing scripts. Use when the user wants to prototype, wireframe, map a user flow, or plan/run usability testing.
Map this token system to or from any external design system (Material Design 3, Apple HIG, Fluent, Carbon, Ant, shadcn/ui, Radix, Chakra, Mantine, Bootstrap…) — adopt their look, build on their stack, or migrate between systems. Use when the user mentions interop, migration, or a specific design-system/component-library bridge.
Turn a reference image, screenshot, or mockup into token-driven, accessible code — infer the design system from the reference (palette, type scale, spacing, radius, layout archetype), map it to the 3-tier tokens, rebuild it, then verify with the kit's gates. Use when the user provides a design/screenshot and wants matching UI code.
Set up or run the token build pipeline — transform the DTCG tokens/*.json (source of truth) into platform artifacts (CSS variables, Tailwind @theme, JS/TS, iOS Asset Catalog, Android, Compose) with Style Dictionary / Tokens Studio / W3C DTCG export. Use when the user wants to generate platform theme files from tokens, wire token CI, or multi-platform token output.
Upgrade an existing website or app to premium quality without breaking functionality — audit the current design, identify generic/AI tells, then apply taste and system rules surgically. Use when the user wants to improve, modernize, polish, or "make better" an existing UI/codebase.
Write or review UI copy — buttons, errors, empty states, microcopy, notifications, labels — using the voice & tone system (clear, concise, useful, human, honest) with the what→why→how error formula and inclusive-language rules. Use when the user needs interface copy, error messages, empty-state text, or a copy review.
Core ComfyUI knowledge — workflow format, node types, pipeline patterns, and MCP tool usage
Authoring ComfyUI v2 frontend extensions with @comfyorg/extension-api — defineNode/defineExtension/defineWidget, shell UI (sidebar tabs, commands, hotkeys), typed events, and handles. Use when writing or editing ComfyUI web-UI extension code (custom node JS, sidebar panels, widgets).
Discover Civitai models with the BUILT-IN search_civitai_models tool and install/generate them locally — find a checkpoint/LoRA/embedding on Civitai, download it into ComfyUI, and use its trigger words. Optionally pair the official Civitai MCP for community features (images browsing, posting, collections).
Anime/illustration text-to-image (ANIMA 1.0, ~2B Cosmos DiT) — use for anime, manga, illustrated characters; accepts Danbooru tags + natural language; runs/trains on <6GB VRAM; includes anime inpainting via Anima-LLLite ControlNet
Diagnose and fix video/image color OBJECTIVELY with the analyze_color tool (scopes/stats — black/white points, contrast, saturation, clipping, cast) instead of eyeballing a contact sheet. Covers the "washed out" signature, why reference color-match (mkl/ColorMatch/ColorMatchAdobe) CAN'T add contrast a flat source lacks, the levels/contrast-stretch fix (core AdjustContrast / CurveEditor), the measure→fix→re-measure loop, the side-by-side sandbox pattern, and where to place the fix in a render graph (after decode, before save). Use when a render looks washed out / flat / dull / over-saturated / color-cast, or when deciding between a color-match and a contrast/levels fix.
Train custom LoRAs with ostris AI-Toolkit — covers WAN 2.2/2.1 (people, styles, video motion) and Z-Image (Turbo & Base, low-VRAM image LoRAs). Use when the user wants to train a WAN or Z-Image LoRA; covers local + RunPod setup, dataset prep, key params, and using the result in a ComfyUI workflow.
Pick the right ComfyUI startup flags for VRAM, attention, caching, and speed — the full decision matrix for OOM (--novram / --cache-none / --disable-smart-memory), shared-VRAM creep on Windows (--reserve-vram N), model-switching with big text encoders (--cache-none), high-VRAM throughput (--gpu-only / --highvram), and attention-backend selection (--use-sage-attention for speed, --use-pytorch-cross-attention as the highest-quality / Z-Image-safe fallback). Also the acceleration-stack + Blackwell/RTX 5000 (sm_120) notes. Use when a graph OOMs (especially long video like LTX 2 / WAN), when the GPU spills into shared VRAM and slows to a crawl, when switching between models eats all RAM, when Z-Image produces black/garbled output under Sage, or when deciding which attention backend to launch with. Flag names verified against upstream comfy/cli_args.py — see Sources.
Train a custom anime LoRA on the ANIMA base model — Citron's local Gradio trainer (kohya sd-scripts), <6GB VRAM, character/style LoRAs; covers setup, dataset prep, training params, and using the result in the anima-base workflow
Build Flux txt2img workflows — Flux.1 Dev (SRPO), Flux 2 Klein 9B, Turbo LoRAs, FluxGuidance, and DualCLIPLoader patterns
Full production pipeline — story to scenes, Z-Image start frames, Qwen Edit end frames, WAN FLF video clips, ffmpeg concatenation
Debug a WRONG or imperfect render (not a hard error) by inspecting inputs and intermediate steps with run-to-node — render one branch up to an output, preview-tap latents/masks/preprocessor maps, localize the first bad stage, then fix. Use when a final image/video comes out wrong — artifacts, wrong subject/pose/composition/color, blur, a ControlNet/IPAdapter/mask/LoRA not taking, a two-stage refiner or upscale degrading the result — rather than the run failing with an error (for errors/OOM/missing nodes use the troubleshooting skill).
Use when installing a model family from an installer pack, or when building/deriving a new pack from an upstream installer or a workflow JSON. Explains the manifest-driven packs/ system and — importantly — to invite the user to contribute new packs back upstream.
Build Ideogram 4 (Ideogram Ultra) txt2img and img2img workflows — local open-weights model, dual conditional/unconditional models with DualModelGuider, Qwen3-VL text encoder, and structured JSON ("compositional deconstruction") prompts for strong text rendering and layout control
Authoring & publishing ComfyUI custom nodes to the Comfy Registry — node structure, pyproject.toml spec, comfy-cli publishing, and CI
Build Baidu ERNIE-Image / ERNIE-Image-Turbo workflows — primarily TEXT-TO-IMAGE. Pick ERNIE when you need precise multilingual text rendering, posters/signage, manga/anime multi-panel layouts, or strong instruction following for complex multi-object scenes. Also supports denoise-based image-to-image refine (NOT instruction-grounded editing — use Qwen-Image-Edit or Flux Kontext for "change X in this photo" edits).
Build Krea 2 Turbo txt2img workflows — native krea2 CLIPLoader, Qwen3-VL encoder, Qwen image VAE, 8-step turbo settings, and Ideogram-style JSON prompting
Run the ComfyUI agent locally for FREE — no subscription, no API key, fully offline — using our gemma4 models fine-tuned on the comfyui-mcp tool suite via Ollama. Use when the user asks about running locally, running for free, offline use, avoiding API costs, Ollama setup, or which local model to pick.
Build Lightricks LTX-2 / LTX-2.3 video workflows — text-to-video, image-to-video, GGUF and bundled checkpoints, distilled model, camera control LoRAs, synchronized audio, two-stage upscaling, and swapping alternate/GGUF base models
Curated download URLs and target directories for every model the comfyui-mcp skills reference — checkpoints, VAEs, text encoders, LoRAs — organized by family (Flux, WAN, LTX, Qwen, Z-Image, SD15/SDXL). Use when downloading models with download_model / download_civitai_model, when a workflow fails with a missing-model error, or when setting up a new machine.
Claude Skills — questions
Answers built from the skills we actually parsed.