The catalog
608,352 rows, by stacks then stars
medic/cht-ai-tools/skills/cht-specialist Skill
1★ repoExpert assistance with Community Health Toolkit (CHT) development, configuration, troubleshooting, and architecture. Use when working with CHT applications, forms (XLSForm/XForm), tasks.js, targets.js, contact-summary.templated.js, app_settings.json, purge.js, cht-conf CLI, Docker deployment, workflows, integrations, CouchDB/PouchDB, offline-first architecture, replication, SMS messaging, contact hierarchy, user roles and permissions, or any CHT-related questions. Provides guidance on building CHT apps, debugging issues, understanding architecture, data models, and following best practices.
Nachi-Kulkarni/Callsmith_skills/skills/callsmith Skill
1★ repoUse when a user is building, changing, reviewing, debugging, testing, or shipping software that listens and speaks: phone agents, WebRTC or app voice, voice notes, and STT-to-LLM-to-TTS or realtime speech stacks. Also trigger when such a system waits for the caller to speak before greeting, has slow first audio, broken interruption or echo, audio-format mismatches, unsafe consent, retention, or handoff, uncertain provider or architecture choices, deploy drops, multilingual failures, or unproven scale and cost claims. Turn the brief or code into pack-backed decisions, enforced safety floors, measured audio behavior, a validated handoff contract, and framework-native implementation.
adraalabs/ios2android/skills/ios2android Skill
1★ repoAudit, implement, visually compare, and continuously update native Android parity from an existing iOS app, treating the current iOS product as the source of truth. Use when an AI coding agent such as Codex or Claude Code must port or mirror SwiftUI/UIKit screens into Kotlin/Jetpack Compose, detect and reproduce iOS commits that landed after the last verified Android sync, generate paired screenshot overlays, heatmaps, contact sheets, and reports, compare layouts and UI states, align feature/file structure, resolve cross-platform drift, or build and verify both apps in simulators/emulators.
AIBIO-UK/airbds-dev/skills/GF/GF-airbds-assessment-skill Skill
no license1★ repoDevelopment repository for the AIRBDS metric project
Strapazzon/skill-nextjs-starter/skills/nextjs-starter Skill
no license1★ repoClaude Code skill for Next.js starter projects
daizhouchen/market-research-skill/skills/market-research Skill
1★ repoAdaptive market demand analysis skill for Claude Code - 自适应市场需求分析 Skill
vordimous/mikado-skills/plugins/mikado/skills/mikado-loop Skill
1★ repoHariSeldon86/skills/skills/simulink-creator Skill
1★ repojagypus/systemic-investing-design/skills/systemic-investing-design Skill
1★ repoDesign systemic investing strategies, portfolios, and vehicles — capital deployed to transform human and natural systems, not to fund single-point solutions deal by deal. Integrates TransCap (system→capital reversal, strategic portfolios, combinatorial effects, backbones), TWIST wealth-holder practice, Tamarack’s portfolio typology, catalytic/blended capital, and developmental + Blue Marble evaluation. Use when a user designs or assesses capital aimed at systems change — wealth holders investing for systems change, foundations asking if their impact portfolio is systemic, fund designers structuring blended vehicles, place-based portfolios, capital mixes across grants/concessional/market-rate, or systems-portfolio measurement. Trigger broadly on "impact portfolio", "blended finance", "catalytic capital", "place-based investing", "portfolio approach", even without "systemic". Not for designing one venture’s own ownership, governance, or financing from the venture side (use regenerative-business-design).
BackGwa/Hydrogen/skills/hydrogen Skill
1★ repoHydrogen, Agent development principles that reduce guesswork, trim overengineering, and leave only the changes that matter.
lucaspedrozaem/saasskills/skills/wireframes Skill
1★ repoWhen the user wants to wireframe a SaaS application page or feature. Use when the user says "wireframe," "sketch," "layout," "page structure," "wireframe this feature," "rough design," "lo-fi mockup," or needs to plan the visual structure of any screen before moving to high-fidelity design. For visual polish, see ui-design. For user journeys, see ux-flows. For interactive demos, see prototype. For component standards, see design-system.
morris-y/blueprint-skills/skills/blueprint-plan Skill
1★ repoUse when transforming a new trading strategy idea into actionable Linear tickets. Supports Ralph Loop for thorough context gathering. Triggers include strategy brainstorming, feature planning for Polymarket, or when user says "plan a new strategy". Includes mandatory Postmortem immune screening.
marinvch/Cortex/skills/writing-for-agents Skill
1★ repoPortable GitHub Copilot context engine — scan any repo, generate agents, skills, MCP tools, and slash commands
a3lem/shablon/skills/use-shablon Skill
no license1★ repoA simple tool for generating project files from Jinja2 templates
cskwork/streamdeck-agent-cockpit/.cursor/skills/streamdeck-agent-cockpit Skill
1★ repoTurn a Stream Deck into a standalone local cockpit for Claude Code, Codex, Pi, JCode, and other terminal agents. No MCP server, no cloud.
krzemienski/multi-agent-consensus/skills/consensus-pipeline Skill
1★ repoFramework for 3-agent consensus validation with hard gates for Claude Code
ophis-fi/skills/plugins/ophis/skills/ophis-swap Skill
1★ repoOfficial Ophis plugins for Claude Code: swap tokens onchain via the Ophis MCP server (MEV-protected, gasless, keyless).
The same file is in 1 other repository
lorismaz/charm-ruby-claude-plugin/skills/charm-ruby-development Skill
no license1★ repoClaude Code plugin for building beautiful CLI tools in Ruby using the charm-ruby ecosystem
The same file is in 1 other repository
KensioSoftware/kensio.ai/plugins/isolated-testing-style/skills/isolated-testing-style Skill
1★ repoKensio Software AI skills and tooling
Galtea-AI/skills/skills/galtea-loop Skill
1★ repoRun the full Galtea improvement loop for an AI product -- define product specifications, generate test cases from them, evaluate the product against those datasets, then iterate the product (and the specs) based on which test cases failed. Use when the user wants to measurably improve an AI agent/product with Galtea end to end, set up a repeatable eval-and-iterate cycle, or asks "how do I make my agent better with Galtea" rather than a single one-off Galtea call. Vendor-neutral: works when driven by Claude Code, Codex, Cursor, or any Agent-Skills-compatible coding agent, using either the `galtea` CLI or the Python SDK.
rhys117/claude-rulekit/plugins/rulekit/skills/sorbet-inline-rbs Skill
1★ repoRead & Write rules for providing context just in time to claude code
Mapika/paper-reader/plugins/paper-reader/skills/paper-reader Skill
1★ repoClaude Code plugin. Fetches research papers (PDF + arXiv tex source) and a real BibTeX entry from open sources (OpenAlex, Crossref, DBLP, ACL, Semantic Scholar). Every claim cites a section or table/figure plus page. +23pt pass rate vs no-skill baseline.
dhanush1348/Motion-Patterns/skills/motion-patterns Skill
no license1★ repoProduction-ready animation patterns for React / Next.js — button, modal, toast, stagger, page transitions, exit animations, scroll, and layout — built on motion-foundations tokens and springs.
giovi321/obsidian-wiki/skills/wiki-setup Skill
1★ repoInterview schema, template substitution rules, and registry write protocol for the /setup-wiki command. Defines the standard entry-point suggestions, the substitution map for CLAUDE.md and template files, and the validation rules applied before writing to disk.
ayushnangia/alliance-ml-docs/skills/alliance-ml Skill
no license1★ repoAlliance Canada HPC documentation for ML researchers — scraped docs, llms.txt files, and Claude Code skill
Eclipse-Space/ast-cli/skills/ast Skill
no license1★ repoEclipse Agent Studio CLI — install the `ast` binary + agent skills (Claude Code / Gemini / Codex)
DIMO-Network/dimo-platform-skill/plugins/dimo-platform/skills/dimo-platform Skill
no license1★ repoDIMO developer platform knowledge for Claude Code: APIs, SDKs, SACD permissions, on-chain protocol + verified contract addresses.
hiivmind/plugin-portability/skills/plugin-portability Skill
1★ repoA plugin for Claude Code, Cursor, Gemini CLI, Codex that analyzes any plugin for cross-platform portability gaps and creates missing platform artifacts. No CLI, no scripts, no framework. Just skills that help you author your skills and plugins for multiple agentic tools
Fabio-RibeiroB/tui-designer/plugins/tui-designer/skills/tui-designer Skill
1★ repoA Claude Code plugin that helps design Terminal UIs — layout, keyboard interactions, component choices, and tech stack.
ofekamirav/kmp-clean-architecture-skill/plugins/kmp-clean-architecture/skills/kmp-clean-architecture Skill
1★ repoKMP + Clean Architecture skill for Claude Code, Cursor, Copilot, and other AI assistants. Scaffolds shared modules with Koin/Ktor/SqlDelight/Firebase/SKIE. Battle-tested patterns.
Recreate scientific figure in visio
TZ-mx/Visio-Illustrator/plugins/visio-scientific-illustrator/skills/recreate-scientific-figure-in-visio Skill
1★ repocho1124/multi-agent-adversarial-verification/plugins/adversarial-verify/skills/adversarial-verify Skill
1★ repo멀티 에이전트 적대적 검증 시스템 — 변증법 기반 AI 에이전트 오케스트레이션
Shipu/service-documenter/skills/service-documenter Skill
no license1★ repoGenerate structured docs/ for any microservice — polyglot, any framework, any database.
dnvriend/markdown-quiz-exporter-tool/plugins/markdown-quiz-exporter-tool/skills/markdown-quiz-exporter-tool Skill
1★ repoA CLI tool that exports quiz markdown files to Anki, Flashcard Hero, and interactive HTML
The same file is in 2 other repositories
sangrokjung/loop-forge/skills/loop-forge Skill
1★ repoTurn a one-line description of a repetitive task into a reusable, self-guarding slash command. loop-forge diagnoses the task into one of 5 loop shapes (Batch / Pipeline / Refine / Watch / Explore), interviews for the blanks, and auto-injects two safety devices the user didn't know they needed — an independent verifier (maker ≠ checker) and a hardstop (a budget/count/cooldown ceiling) — then previews the result and stamps it as a `/command` they can run forever. Use when the user says "/loop", "/loop-forge", "/make-it-loop", "automate this", "make this repeatable", "turn this into a command", "do this for all 100 items", "do X every time Y happens", "generate several and pick the best", or otherwise wants to capture a recurring task as a reusable slash command instead of re-typing the prompt by hand. Works in any language: it interviews the user and writes the stamped command in the user's own language. Non-goal — it does not schedule unattended runs (launchd/cron) or publish externally on its own; it stamps the reusable command and stops there.
adelaidasofia/claude-insights/skills/insights Skill
1★ repoWeekly and monthly journal pattern recognition for Claude Code. Floor trends, life coach observations, advisory panel. Companion to claude-daily-journal.
averliz/visual-ux-review-toolkit/skills/ux-flow-walkthrough Skill
1★ repoJourney-level UX walkthrough — drives a LIVE app as a clueless first-time user, one click at a time, to judge whether the flow actually makes sense. Runs an automated cognitive walkthrough: at each step it reasons as a naive user (do I know what to do? is the right control obvious? does its label make sense? is there feedback?), takes the most-obvious action, and logs every hesitation, dead-end, missing feedback, confusing label, unexpected navigation, or trap. Three modes: goal-directed (attempt a named task like 'sign up' or 'create a post'), free-explore (wander from an entry URL with no goal), and full auto-explore (exhaustively + adversarially traverse all reachable flows). Produces a step-by-step journey report with friction flagged by severity and a 'can a real first-timer actually complete this?' verdict. Use whenever the user wants to test a flow/journey/user-path or sense-check the experience — phrasings like 'does this flow make sense', 'walk through the app as a user', 'test the signup/checkout/onboarding flow', 'click through it like a confused user', 'find the UX dead-ends', 'is this confusing for a first-timer', or as a review step after building a feature to validate the end-to-end experience. Complements visual-ux-review (which checks how each SCREEN looks); this checks whether the FLOW between screens works. NOT for per-screen visual/pixel critique (use visual-ux-review), code/PR review, writing automated test scripts, or implementing features — only for experiencing and judging a live user journey.
juliuszfedyk/dev-journal/skills/journal Skill
1★ repoDevelopment journal: write, search, and retrieve project decisions and session logs. USE WHEN: user says "log", "journal", "write entry", "record what we did", "what did we decide about", "search journal", "show last entries", "check context", "what have we done". USE PROACTIVELY: invoke /journal check at the start of any session or before any architectural decision (new file, API design, protocol choice) without being asked. DO NOT USE for: general file writing, git commits, or inline code comments.
codefoundry-io/skill-prompt-review-claude/skills/skill-prompt-review Skill
1★ repoDiagnoses a SKILL.md or an authored prompt (a system prompt, or a worker/dispatch prompt written for a CLI such as codex, gemini, or claude) against skill- and prompt-authoring best practices, returning a per-criterion PASS / FAIL / N-A report with evidence and a concrete fix. Use when authoring or revising a skill or a worker prompt, before shipping one, or when a skill under- or over-triggers or a prompt underperforms. For a sub-agent definition file (an agent `.md`), review it as an agent definition, not with this skill.
The same file is in 1 other repository
longkeyy/claude-discover/skills/content-discovery Skill
no license1★ repo🔍 AI-powered multi-source content discovery plugin for Claude Code. Discover, analyze, and publish quality content from ArXiv, GitHub, HuggingFace with semantic deduplication.
The same file is in 2 other repositories
taxbit-shared/taxbit-skills/skills/utilities Skill
no license1★ repochussum/cat-harness/skills/team Skill
1★ repo🐱 Ask-before-building workflow harness for Claude Code — auto-routes vague requests through interview → consensus plan → evidence-gated execution. No slash commands needed.
haplesshero13/agent-trio/skills/using-agent-trio Skill
no license1★ repoNatural-language scaffolding for agentic coding that minimizes regret while maintaining autonomy
vindu939/gh-file-attach-plugin/skills/file-attach Skill
1★ repoA Claude Code plugin that lets Claude upload files to GitHub and get shareable URLs. Replicates GitHub's "Attach files" feature, powered by the official Releases API.
kylesnowschwartz/shelf/skills/shelf-skill Skill
1★ repoKeyless, local reading-recommendation engine + Claude Code plugin
LLaammTTeerr/competitive-programming/skills/writing-statements Skill
1★ repomaneja81/claude-skills/plugins/weekend-project/skills/weekend-project Skill
1★ repoAgent skills for Claude Code — weekend-project (autonomous PR-by-PR project builder) and code-better (40 stackable behavioural mode commands). Personal side project, built evenings and weekends.
stakewise/llm-tools/data-skill/skills/stakewise-data-query Skill
1★ repolukasmalkmus/pngx/skills/paperless Skill
1★ repoAI-native CLI for Paperless-ngx
revans/buster-marketplace/documenter/skills/system-documentation-skill Skill
1★ repoAgents, Skills, Commands, & more
yogvidwankhede/vishwakarma/.claude/skills/public-api-integration Skill
1★ repoUse when an interface needs live third-party data and no API is chosen yet: selecting from public-apis, probing it, and wiring it safely.
bhutano/steelprompt/skills/steelprompt Skill
no license1★ repoUse to manually engineer any prompt before sending it. Applies all official Anthropic prompt engineering principles and rewrites a complete, structured version. Also supports switching modes.
mjtrapani/k8s-rbac-companion/skills/rbac-reference Skill
no license1★ repoUse when working with Kubernetes client code (client-go, controller-runtime, the kubernetes Python client, or kubectl), generating or reviewing Kubernetes RBAC (Role/ClusterRole plus RoleBinding/ClusterRoleBinding), or discussing RBAC rule syntax (apiGroups/resources/verbs/resourceNames). Provides the RBAC rule-grammar primer, the Role-vs-ClusterRole scope map, built-in and aggregated ClusterRoles, and pointers to detailed reference docs (API-resource map, version deltas, client-library patterns, resource extraction).
promptromp/aws-bootstrap-g4dn/aws-bootstrap-skill/skills/aws-bootstrap Skill
1★ repofast and easy bootstrapping of AWS EC2 instances for CUDA development. Use as a CLI, as a programmatic SDK, or as an Agent Skill!
Inco-fhevm/skills/skills/lightning Skill
no license1★ repoInco Skills
The same file is in 1 other repository
carryons6/research-guardrail-skill/skills/research-guardrail Skill
1★ repoGuardrail for AI-assisted research and analytical/computational work in ANY field — data analysis, statistics, machine learning, simulations, modeling, quantitative finance, computational science, and similar. Two jobs: (1) preserve the researcher's understanding and ownership of the parts worth owning, and (2) — the bigger threat in data work — catch the "looks-fine-but-silently-wrong" code (unit/scale mismatches, index/axis conventions, sign and ordering errors, data joins and alignment, normalization, time/epoch handling, leakage) before it quietly corrupts a result. EXPLICIT INVOCATION ONLY — do NOT trigger this skill automatically or proactively. Apply it only when the user explicitly asks for it: by running the /research-guardrail command, saying "$research-guardrail", or naming the skill (e.g. "use the research-guardrail skill to ..."). Do not invoke it on your own even when the task looks like reproducing a result, new-method development, pipeline building, or a quick implementation that could corrupt a downstream conclusion.
hungson175/pdf-to-epub-skill/skills/pdf-to-epub Skill
1★ repoConvert a native-text PDF book/manuscript into a clean, faithful EPUB (headings, paragraphs, dialogue, footnotes, bordered tables, embedded photos) ready for Kindle. Use when Boss wants to read a PDF book on Kindle/e-reader, or asks to "convert this PDF to epub", "turn this book into an ebook", or names this skill directly. Handles books with ruling-line tables (lookup grids, reference charts) as well as continuous prose -- NOT for business documents/decks with charts/graphics-as-data, use document-processor for those.
agent-sh/system-prompt-curator/skills/system-prompt-curator Skill
no license1★ repoBaselightDB/skills/skills/baselight Skill
1★ repoAgent Skills to access the Baselight open data platform
ryannadel/CogMap/.agents/skills/build-cogmap Skill
1★ repoIncrementally re-ingest notes (Markdown, plain text, or optional OneNote .onex) and build/refresh the CogMap Minard-style Knowledge Base visualization (an interactive knowledge-base-viz.html). Use when the user says "build the knowledge base", "refresh the knowledge base", "update the visualization", "re-ingest my notes", "add these notes", "I changed the notes", "rebuild the knowledge map", or points at note files to add. Only the chunks that actually changed are re-extracted. Runs a resumable one-command orchestrator and spawns extraction / resolution / synthesis sub-agents (or runs those stages itself) only when the delta requires it. No API key needed — the host coding agent is the LLM in the loop.