The catalog
610,220 rows, by stacks then stars
rilixai/beaker-skill/skills/beaker-usage Skill
5★ repoPortable Agent Skill for setting up Beaker harness optimization
libovin/cocos-skills/skills/cocos-skills Skill
no license5★ repobitbankinc/bitbank-lab-mcp/skills/investment-onboarding Skill
5★ repo暗号資産投資の進め方をユーザーと一緒に決めるためのスキル。 「何から始めればいい?」「どう運用すれば?」「投資戦略を相談したい」 「暗号資産で増やしたい」「bitbank で運用したい」「ビットコインで稼ぎたい」など、 上流の方針が定まっていないユーザー発話を起点に、リスク許容度・運用スタイル・ 具体的な戦術を最小限の問い(最大 3 問)で絞り込み、該当する戦術ガイド (references/style-*.md)と MCP ツールへ誘導する。 ユーザーが既に運用スタイル(資本保全 / インカム / グロース / アクティブ / レバレッジ)や戦術を明言している場合は、ヒアリングを省略して該当ガイドに 直接バトンする。
Schoepplake/framer-motion-skill/skills/framer-motion Skill
no license5★ repoProduction-grade animation patterns for React and Next.js using Motion (formerly Framer Motion). Use this skill whenever the user asks to animate components, add transitions, create scroll-triggered effects, implement page transitions, layout animations, gesture interactions, or any kind of motion/animation in a React or Next.js project. Also trigger when code imports 'framer-motion', 'motion/react', or 'motion/react-client', or when the user mentions: animation, transition, fade-in, slide, parallax, scroll animation, exit animation, AnimatePresence, motion.div, spring, gesture, drag, hover animation, stagger, whileInView, or layout animation.
contextualizer-ai/cyberian/skills/cyberian-control Skill
5★ repoControl and coordinate multiple Claude Code sessions via cyberian's agentapi wrapper. Use this proactively when tasks involve multi-agent orchestration, delegating work to other agents, or managing agent farms.
lexiaoyao20/excalidraw-generator/skills/excalidraw-generator Skill
5★ repoProfessional Excalidraw diagram generation expert that creates complete, valid .excalidraw files based on user descriptions. Use ONLY when users explicitly mention "Excalidraw" or request to create diagrams specifically with Excalidraw (e.g., "用 Excalidraw 画个流程图", "generate an Excalidraw diagram", "create Excalidraw file"). Do NOT trigger for generic diagram requests without Excalidraw mention. Generates ready-to-open files instead of requiring copy-paste.
aiforging/aiforging/skills/architecture-analyzer Skill
5★ repoUse when /aiforging:setup has confirmed a backend or fullstack project and needs a non-destructive, advisory audit of how closely that project's current structure aligns with the AI Forging architectural ideals. Produces a structured ANALYSIS.md report with a score, findings, and severities. Never modifies files.
hasansezertasan/copier-pyproject/skills/update Skill
5★ repoThis is an opinionated pyproject template.
UltimaPhoenix/dev-coach/plugin/skills/devcoach Skill
5★ repoStay sharp while your AI does the work — local MCP coach that teaches one concept per task.
glennmatlin/gt-hive/skills/slurm-core Skill
5★ repoGeorgia Tech PACE HPC skills for AI coding tools — Claude plugin and Codex skill for navigating Phoenix and ICE clusters.
uberblick-ai/skills/plugins/uberblick/skills/uberblick Skill
5★ reposkills to work with opinionated uberblick product development flow.
ChristianKohlberg/Takomo/plugins/takomo/skills/takomo Skill
5★ repoA hosted, single-source-of-truth task tracker for AI agent fleets — configurable enforced workflows, atomic claims/leases, epics & dependency graphs, with an MCP server, CLI, and web board.
VoxTechnologies/transmute-framework/skills/smoke Skill
5★ repoRuns production smoke tests to verify the deployed application works correctly in the live environment. This skill should be used when the user asks to "run production smoke tests", "verify the production deployment", "smoke test the live app", "check if production is working", "run post-deployment verification", "run stage 7V", or "verify the deployed application", or when the transmute-pipeline agent reaches Stage 7V of the pipeline.
serkutYILDIRIM/cobol-to-java-modernization-skills/plugins/cobol-modernization/skills/diagram-cobol-with-mermaid Skill
5★ reposixtyfour-ai/skills/skills/sixtyfour Skill
5★ repoDefault for any task involving people or company intelligence. Use whenever you need to investigate, research, or gather structured information about a person, company, or organization — whether it's a quick lookup or a deep investigation. Covers identity resolution, competitive intelligence, due diligence, org chart mapping, relationship mapping, risk signal detection, and answering any question about who someone is or what a company does. AI research agents return verified, source-backed answers with no fixed schema — you define the fields you need. More comprehensive and reliable than manual web research for structured people and company data.
Fornace/claude-overnight/plugins/claude-overnight/skills/claude-overnight Skill
5★ repoRun parallel Claude Code agents with real-time terminal UI
lbnl-science-it/scienceit-docs/skills/lawrencium-hpc Skill
no license5★ repoScienceIT Technical Documentation
nikitaCodeSave/claude-code-harness/plugins/devlog/skills/devlog Skill
5★ repoClaude Code harness kit — designs, bootstraps, audits, and extends a project's .claude/ + root CLAUDE.md. Under a capable model, less harness yields more productivity.
yugui923/claude-vc/skills/vc Skill
5★ repoVenture capital analysis toolkit. Give it a URL or pitch deck and it runs screening + memo. Type "/vc help" for details on all 6 commands.
hkasahar/claude-code-skills/skills/gemini-pdf Skill
5★ repoConvert PDF files to clean Markdown using Antigravity CLI (`agy`). Best for academic papers with math, tables, and structured content. The native pipeline feeds page-range chunk PDFs directly to Gemini's vision pathway (agy reads the PDF itself — no lossy text extraction), with an automatic quality-gated fallback to a Mathpix hybrid and Mathpix standalone. Zero marginal cost via a Google subscription. Triggers include "convert PDF to markdown with Antigravity", "agy PDF conversion", "PDF to markdown with math", and "convert paper to markdown".
jkerdels/dew-plugin/skills/dew-develop Skill
no license5★ repoA lightweight claude code plugin for the 6D development workflow.
littlestudent886/complex-dev-skills/claude/plugins/complex-feature-dev/skills/complex-feature-dev Skill
5★ repopengyizhou/claude-plugins/reviewing-my-paper/skills/reviewing-my-paper Skill
5★ repoedbizarro/glaze/skills/Glaze Skill
5★ repoTurn any output you already have — a plan, a report, a summary, release notes, any text — into a beautiful, self-contained themed HTML page. Glaze *skins* content; it does not generate it. Ships an eight-theme library (synthwave · manga · brutalist · terminal · sumi · coffee · dieselpunk · neonops), each a self-contained CSS preset built from award-winning design techniques. Zero build step, no JS framework, fonts via Google Fonts, all CSS inlined so the file travels as a single standalone artifact. Extensible: a new theme is one CSS file. Output is verified by opening it in a browser. USE WHEN glaze, glaze this, render as HTML, themed HTML, style this output, make this pretty, beautify this plan/report/summary, skin this content, apply a theme, turn this into a styled page, --style, synthwave/manga/brutalist/terminal/sumi/coffee/dieselpunk/neonops HTML, show the theme catalog. NOT FOR generating the content itself (produce the content first, then glaze the result), static illustrations or diagrams, or video.
isizono/calm/skills/recompose-context Skill
5★ repoclaude codeさんが作業内容を記録するためのメモリ
submodule-org/submodule.go/skills/submodule-go Skill
5★ repoalexeyv/pan-out/skills/panout-help Skill
5★ repoAI Cooking Assistant
Farfield-Dev/deep-review/plugins/deep-review/skills/run Skill
5★ repoThe bug-finding recipe Farfield uses in production, packaged as a Claude Code skill. Free + MIT.
wayne930242/straw-boss/skills/dispatching-work Skill
5★ repoHerdr-backed task dispatch for Claude Code: run implementation work in a session rooted in your app's own directory, single-app or across a monorepo
gquattromani/prompt-best-practices/skills/prompt-best-practices Skill
no license5★ repoStructure a prompt before executing it. Use this BEFORE acting on any request to write, draft, generate, implement, fix, refactor, build, create, design or analyze something, whenever that request does not state its success criteria, output format or constraints. Also use it when the user asks to improve, optimize, review or rewrite a prompt, or invokes /prompt-best-practices. The assessment is one message and the user can skip it with one word, so activating is cheap. Composes the final prompt for the runtime it will run on (Claude, GPT, Gemini, Grok, or a portable baseline). Not for micro-tasks (one rename, a typo, a single import) or for questions.
az9713/impeccable-tutorial/.gemini/skills/impeccable Skill
5★ repoFork of pbakaus/impeccable — AI frontend design skill with extended documentation
iaserrat/krafter/skills/quality-branch Skill
5★ repoEvidence-backed, deterministic code-quality review of a git branch's changes in whatever repository the skill is invoked from — language and framework agnostic. Quality judgement is subjective, so this skill makes the EVIDENCE deterministic: a bundled Rust tool (cqt) measures per-function complexity deltas (cognitive/cyclomatic), repo-relative percentiles, churn-x-complexity hotspots, branch-introduced duplication, and missing co-change coupling, and joins them with `cqt assess` into ranked, reason-coded candidate findings. Every finding must cite a measurement at file:line. No measurement, no finding. Use whenever the user asks to review code quality, maintainability, complexity, or duplication of a branch/PR/diff, asks "did this branch make the code worse", "what got more complex", "where's the risky code", or wants a reproducible, metric-grounded quality assessment rather than vibes. Produces a ranked, measurement-cited findings report with remediation pointers — it assesses, it does not fix.
Wayfound-AI/wayfound-agent-skills/agent-skills/wayfound/skills/wayfound Skill
5★ repoadewale/good-pr/skills/good-pr Skill
no license5★ repoA Claude Code skill that helps you craft pull requests maintainers actually want to merge
jarfa/critique-loop/plugins/critique-loop/skills/critique-loop Skill
5★ repoA Claude Code plugin that enables two Claude Code instances to collaborate through turn-based dialogues with automatic notifications.
WindyITS/skills/skills/study Skill
5★ repoTeach the user any subject through comprehensive interactive lessons, inside a stateful learning workspace. Mission-driven, evidence-based teaching (missions, learning records, curated sources, glossary) combined with a rigorously specified lesson format — Jupyter notebooks for subjects learned through code and computation, self-contained HTML for everything else. A starting probe maps prior knowledge into a living DAG (MAP.md); every lesson runs in two phases (markdown outline first, then the artifact), and checkpoints gate the exercise blocks where enough new material has accumulated.
GxFn/Wakeflow/core/skills/wakeflow-target-craft Skill
5★ repoDual-host Wakeflow plugins for Codex and Claude Code: controller state roots, task packages, direct-thread/session delivery, evidence review, and local-first runtime.
Asher-pro/wa-whatsapp-agent/skills/wa-characterize Skill
no license5★ repoClaude Code plugin: Build, deploy & maintain a WhatsApp AI agent - no coding required
xiaosongz/sas94-skill/plugins/sas94/skills/sas94 Skill
5★ repoClaude Code skill for SAS 9.4 — DATA step, PROC SQL, macros, hash tables, ODS, stats procs
Brian-1812/marketing-launch-video/skills/marketing-launch-video Skill
5★ repoProduce a professional, animated launch or marketing video for a software, SaaS or AI product — given only a product URL, or a local dev server, or an existing repo. Rebuilds the product UI in Remotion as frame-driven React (rather than screen-recording it), then choreographs camera, typography, sound effects and a music bed into a 30–60s cut. Works even when the product is behind a login, by building a chaptered feature film from the landing page's own feature cards and components. Use this skill whenever someone asks for a launch video, product demo video, promo video, marketing video, product trailer, teaser, "a video for our landing page", "something to post on X/LinkedIn for launch", "an Apple-style / YC-style product demo", or wants to animate their app for marketing — even if they only paste a URL and say "make a video for this". Also use it when someone wants to improve, retime, re-score, add background music or sound to a product video they already have, or asks for a specific musical feel (energetic, calm, cinematic, minimal, warm, tense).
sitebulb/sitebulb-plugin/plugins/sitebulb/skills/sitebulb-release-check Skill
5★ repostoleas/ansible-skills/skills/ansible-cop-review Skill
5★ repoReview Ansible code against Red Hat Communities of Practice automation good practices. Use this skill when the user asks to: "review ansible code", "check CoP compliance", "validate against Red Hat standards", "cop review", "ansible best practices review", "check ansible code quality", "review playbook", "review role", or wants to validate their Ansible automation against Red Hat CoP standards. Always invoke this skill for code review and validation tasks.
kennyzheng-builds/agentslink/skills/agents-link Skill
no license5★ repoAgent-link project
tikeda/design-extraction-skill/skills/design-variants Skill
no license5★ repoClaude Code skill: Extract design tokens and components from repository code and build a design system in Pencil (.pen) files
ladamczyk-it/qoq/skills/qoq Skill
5★ repoQuality toolkit for JavaScript/TypeScript repos — `fix`, `refactor`, `bump`, `plan`, `execute`, `test`, `compress`. Use it whenever the user wants lint or formatting cleaned up, dead code or dead dependencies removed, a branch checked before merge, duplication refactored, npm dependencies bumped, a spec broken into tickets, an approved plan executed or resumed, tests written for code that already exists, or agent-facing markdown (CLAUDE.md, AGENTS.md, skill docs) made terser to stop burning context. Trigger it even when the user never says "qoq" or names a tool.
SAKURAfan1023/aigc-skill/.claude/skills/aigc Skill
5★ repo学术写作与 AIGC 降重 skill,适用于 Claude Code为首的多种AI工具
build-plus/skills/.claude/skills/bid-leveling Skill
no license5★ repoAnalyze and compare construction bids, proposals, estimates, and subcontractor quotes. Assigns line items to UniFormat II headings, compares absolute cost and cost per unit, produces a leveling spreadsheet (CSV or Excel) and a concise PDF report with scope overlap matrix, contractual term comparison, and a recommendation. Use when the user provides one or more bid documents (PDFs, spreadsheets, or text) to analyze or compare, or asks to "level bids", "compare bids", "bid tabulation", "bid comparison", "bid analysis", or "analyze this bid/quote/proposal" for a construction project.
Repello-AI/artemis-recon/skills/mentor-mode Skill
5★ repoUse when the user starts or runs an authorized AI red-team engagement focused on the AI assets of an application or product — chatbots, agentic features, LLM-powered endpoints, RAG / vector search, AI-assisted UX. The engagement is adaptive multi-turn reconnaissance: legitimate-conversation probing across four priority classes (system-prompt observation, tool inventory, parroting / competitor handling, AI identity). Activates the Mentor controller; Claude defers to MENTOR directives from the `mentor` MCP server for session-steering.
tmoody1973/pasteup/skills/pasteup Skill
5★ repoPipeline for halftone paper-collage and stop-motion-graphic ads. Decodes a reference image into a field-editable JSON spec, generates cut-out elements directly on the Gemini image API (nano-banana line), then assembles them into build-from-empty stop motion with code: a bundled Python compositor by default, or OpenMontage's HyperFrames (HTML + GSAP) or Remotion runtimes. No paid video model needed; the assembly is deterministic. Trigger whenever the user pastes a reference to decode, replicate, or make "similar"; asks for a halftone, paper-cut, paper-collage, editorial-collage, or stop-motion-graphic look; wants a still or product turned into an animated ad or motion graphic; or says "analyze this", "break this down", "animate this", "make a collage ad", or "assemble it in". Also trigger for guided creation: "director mode", "guide me", "interview me", or a vague ask like "make me something cool for my brand". Use even when the user never says "collage" or "stop motion" but the visual is clearly this look.
semnil/urx-router/.claude/skills/urx-routing-planner Skill
5★ repoUse this skill ANY time the user mentions a YAMAHA URX22, URX44, or URX44V (also written urx22 / urx44 / urx44v) in the context of audio signal flow — routing, patching, mixes, sends, channels, buses, USB outputs, DAW tracks, headphones, monitor mix, streaming mix, FX/reverb sends, pre/post-fader, AUX, mic inputs, duckers, or microSD recording. This includes: feasibility questions ("can the URX… / is it possible to… / how do I…"), setup/design requests ("build me a routing plan", "I want to send X to Y"), troubleshooting a routing idea, and any vague "I just got a URX, how do I…" question. Works in any language (English, Japanese, etc.). The skill answers feasibility from a bundled per-model route table, and for settled requests emits a validated URX Router plan JSON plus a ?plan= deep link. Do NOT use for other Yamaha consoles (TF, QL), other interfaces (Focusrite, MOTU), Voicemeeter, DAW routing, modular synths, or generic audio-routing theory questions that don't mention a URX model.
sibucho86/framework-consultant/skills/framework-consultant Skill
5★ repo87種の思考フレームワークを横断選定・適用・統合して新しい視点を返す Claude Code プラグイン / Framework consulting agent plugin for Claude Code
Shimin-Zhang/inhabited-design/skills/inhabited-design Skill
5★ repoBuilds polished, target-user-rated web apps/pages/tools/assets via Verbalized Sampling at every input — framing, designer voice, seeds, constraint, competitors, typography. Use when you want every design choice to trace to a deterministic random sample drawn from an enumerated pool rather than model reach. Triggers on phrases like "iterate to a perfect score," "no canon," "really diversify the inspiration," "use IFG/VS," or explicit "use inhabited-design."
derekherman/walter/.claude/skills/walter Skill
5★ repoThe engineering discipline AI agents are missing. No half measures.
zipchat-ai/zipchat-custom-tool-skill/skills/zipchat-custom-tool Skill
5★ repodwgx/claude-codex-subagent/skills/codex-subagent Skill
5★ repoDelegate expensive work to the local Codex CLI (`codex exec`) as a worker subagent to save Claude's token budget. Use this skill whenever a task involves web search / online lookup, reading or scanning large files, bulk code analysis, writing long files, running tests or scripts, or any chore that would otherwise burn a lot of Claude context. Claude stays the orchestrator — planning, deciding, verifying — while Codex does the heavy I/O-bound or context-heavy work. Trigger on phrases like "用 codex", "丟給 codex", "讓 codex 查", "codex 搜一下", "codex 分析", "dispatch to codex", "delegate to codex", whenever the user hints they want to conserve Claude's tokens or offload work, or whenever Claude is stuck and wants a second pass with a fresh context.
justinstimatze/hybrid/skills/hybrid-loops Skill
5★ repoFind and design hybrid-loop surfaces in any project — places where an LLM's fuzzy semantic judgment is consequential enough to warrant typed structure (schema'd substrate, deterministic gating, calibration). Triggers on prompts like "build a tool that tracks/analyzes/evaluates/extracts X over time", "make sense of Y across many Zs", "detect patterns in W", "notice when X is happening", "score/rank/compare documents", "monitor my agent for X", "watch this stream and alert when...", "build a regression detector for...", "log and analyze evaluator outputs over time", "suggests when I'm repeating an anti-pattern", "flag for me when...", "build a system that learns from outcomes", or any project where the value is DATA ABOUT content rather than the LLM's reply. Diagnostic-first — most projects have only 0-3 such surfaces. Domain-agnostic. Do NOT trigger for one-shot transforms (translate, summarize, format), pure UIs, CRUD without judgment, code refactors, chatbots, agent harness setup, or generic prompt engineering.
sartrus/modelling-team-skill/skills/modelling-team Skill
no license5★ repoBuild Excel financial models using a three-agent team (Architect, Coder, Challenger). Use when the user asks to build a financial model, Excel model, DCF, LBO, P&L, valuation, projections, forecast, or any quantitative spreadsheet with assumptions and scenarios.
Dynatrace kpi dashboard generator
SudoSmitty/dynatrace-kpi-dashboard-generator/skills/dynatrace-kpi-dashboard-generator Skill
no license5★ repoGenerate a Dynatrace Gen 3 **KPI dashboard** (15–20 business KPIs, required map tile, branded section dividers) and a matching 30‑minute BizEvents injector for a named company, then deploy both via `dtctl`. Use this skill ONLY when the user explicitly asks for a Dynatrace KPI dashboard, business-event KPI demo, BizEvents injector, or a "KPI dashboard for <company>" — do NOT use for generic Dynatrace dashboards (SRE, infra, k8s, services, RUM) or for editing existing non-KPI dashboards. Triggers include phrases like "generate a KPI dashboard", "build a BizEvents demo for <company>", "spin up a KPI dashboard + injector", "/generate-kpi-dashboard". Requires `dtctl` authenticated to a Dynatrace Gen 3 tenant.
The same file is in 2 other repositories
EdwardAngert/docs-agent-plugin/skills/docs-assist Skill
5★ repo[WIP] Claude Code skill to help enable technical writers and contributors
kamens/split-plugin/plugins/split/skills/split Skill
no license5★ repoClaude Code plugin for splitting into multiple famous, opinionated personalities to discuss your work.