abubakarsiddik31/claude-skills-collection

A curated collection of official and community-built Claude Skills – extend Anthropic's Claude with powerful, modular capabilities for productivity, creativity, coding, and more.

1,086

61 commits

updated Sep 12, 2026

See the code

README

📦 Claude Skills Collection

A curated collection of official and community-built Claude Skills.
Anthropic Skills are modular tools that extend Claude with practical workflows for coding, debugging, planning, writing, research, design, and automation.

This repository gathers and organizes all publicly available Claude Skills, including both built-in tools by Anthropic and creative contributions from the community. Browse by category, explore capabilities, and kickstart your own Skill creation.

💡 Note: Skills require Claude Pro, Max, Team, or Enterprise access with code execution enabled.

If you use skills across more than one coding agent, Skill Manager gives you one dashboard to discover, enable, disable, edit, and delete them across Claude Code, Codex, Cursor, and OpenCode. Download it here; view the source on GitHub.

Skill Manager — every skill, every agent, one dashboard

📈 Overview

252 skills across 13 categories:

CategorySkills
💻 Development & Code Tools82
📣 Marketing & SEO22
📝 Writing & Research22
🤝 Collaboration & Project Management21
⚙️ Utility & Automation30
🔐 Security & Testing15
📚 Learning & Knowledge15
🎨 Creative & Design13
💼 Career & Job Search6
🎥 Media & Content10
🔬 Scientific & Research Tools5
📊 Data & Analysis6
📄 Document Skills5

These are commonly requested, high-impact skills to start with:

  • grill-me - pressure-test plans and assumptions before execution
  • grill-with-docs - structured planning with CONTEXT.md and ADR updates
  • tdd - red/green/refactor workflow for safer implementation
  • diagnosing-bugs - disciplined root-cause debugging loop
  • handoff - compact transfer notes between sessions/agents

📚 Table of Contents


What Are Claude Skills?

Claude Skills are specialized modules that Claude can use to perform structured, multi-step tasks. Each skill is a lightweight folder with a clear instruction interface, and optionally includes Python code, templates, and assets.

Skills allow Claude to:

  • Create and edit documents (Word, Excel, PPT)
  • Parse and analyze structured data
  • Write and debug code
  • Generate creative visual assets
  • Automate research, testing, or collaboration tasks

Official Skills are created by Anthropic and auto-invoked when needed. You can also install or build custom skills to meet your unique workflow needs.


📄 Document Skills

NameDescriptionLink
docxCreate and edit Microsoft Word documents with formatting, comments, and tracked changesSource
pdfExtract content from PDFs, split/merge documents, or create new onesSource
pptxGenerate and edit PowerPoint presentationsSource
xlsxManipulate Excel files, formulas, tables, and chartsSource
revealjs-skillGenerate polished, professional presentations using the Reveal.js HTML presentation frameworkSource

🎨 Creative & Design

NameDescriptionLink
algorithmic-artCreate generative art using p5.jsSource
canvas-designRender layout-based visual designs in PNG/PDFSource
slack-gif-creatorGenerate Slack-optimized animated GIFsSource
brand-guidelinesApply company branding to outputsSource
theme-factoryCreate and apply visual themes for documentsSource
nano-banana-image-generationCreate images using Nano Banana ProSource
frontend-slidesCreate animation-rich single-file HTML presentations with 12 visual styles and zero dependenciesSource
web-asset-generatorGenerate favicons, app icons, PWA manifests, and Open Graph images with WCAG contrast validationSource
color-expertColor science and accessibility expertise covering APCA/WCAG, palette systems, and color-space decisionsSource
design-mobile-appsDesign mobile app UI screens and implement Sleek designs in HTML, React Native, or SwiftUISource
frontend-designOfficial Anthropic guidance for distinctive, intentional visual design — aesthetic direction, typography, and avoiding templated defaultsSource
web-design-guidelinesOfficial Vercel skill that reviews UI code against Web Interface Guidelines for accessibility, interaction, and UX best practicesSource
emil-design-engEmil Kowalski's design-engineering philosophy: UI polish, component design, animation decisions, and the invisible details that make software feel greatSource

💻 Development & Code Tools

NameDescriptionLink
code-reviewTwo-axis review of changes since a fixed point — Standards (coding standards) and Spec (originating issue/PRD) — run in parallel sub-agentsSource
claude-apiOfficial Anthropic guidance for building, debugging, and optimizing Claude API applications, including tool use, prompt caching, batches, files, citations, and memorySource
claude-skills-pro15 engineering-workflow skills (7-dimension code review, root-cause bug hunting, test generation with mutation check, zero-downtime DB migrations, security audit) — 7 free under MIT, CN handbook includedSource
codebase-designShared vocabulary for designing deep modules — lots of behaviour behind a small interface at a clean seamSource
diagnosing-bugsDisciplined diagnosis loop for hard bugs and performance regressions: reproduce, minimize, hypothesize, instrument, fix, and regression-testSource
domain-modelingBuild and sharpen a project's domain model with ubiquitous language, glossary entries, and architectural decision recordsSource
grill-with-docsGrilling workflow that pressure-tests plans against domain language and updates CONTEXT.md and ADRsSource
implementImplement a piece of work based on a spec or set of tickets, driving TDD internally per issueSource
improve-codebase-architectureFind deepening opportunities in a codebase using shared domain language and architectural decisionsSource
prototypeBuild throwaway prototypes for logic or UI exploration before production implementationSource
researchDelegate reading legwork to a background agent that investigates against primary sources and leaves a cited Markdown fileSource
resolving-merge-conflictsResolve in-progress git merge/rebase conflicts by understanding both changes' intent and preserving itSource
setup-matt-pocock-skillsOne-time repository setup for issue tracker, triage labels, and docs layout used by the engineering skill suiteSource
tddTest-driven development workflow using a red-green-refactor loop and vertical slicesSource
to-specTurn the current conversation into a spec and publish it to the project issue tracker — synthesis, no interviewSource
to-ticketsBreak a plan, spec, or conversation into tracer-bullet tickets, each declaring its blocking edgesSource
triageTriage work through a structured issue-state workflow with explicit rolesSource
wayfinderPlan huge multi-session work as a shared map of investigation tickets, resolving them one at a time until the route is clearSource
interview-meInteractive requirements interview loop to uncover hidden assumptions before implementationSource
source-driven-developmentGround implementation in primary sources and existing code reality before writing changesSource
performance-optimizationDiagnose and improve runtime, render, and bundle performance with systematic profilingSource
code-review-and-qualityEnforce production-grade review criteria, maintainability, and quality gatesSource
incremental-implementationShip complex features as small validated slices to reduce risk and regressionsSource
planning-and-task-breakdownConvert broad goals into sequenced, executable engineering tasksSource
swiftui-proAdvanced SwiftUI development patterns and implementation guidance for Apple platformsSource
swift-concurrency-proPractical Swift Concurrency patterns for structured async code and correctnessSource
react-components-stitchGenerate robust React components using Stitch-oriented implementation patternsSource
MCP ServerBuild Claude-compatible API connectorsSource
Changelog GeneratorCreate changelogs from commit historySource
using-git-worktreesManage feature branches safely in isolated Git worktreesSource
test-driven-developmentWrite tests before implementation to drive developmentSource
subagent-driven-developmentUse multiple Claude subagents to coordinate complex implementationsSource
executing-plansExecute structured plans with checkpoints and verification stepsSource
finishing-a-development-branchComplete development branches with testing and review flowSource
preserving-productive-tensionsManage architectural decisions by preserving competing viewpoints to balance innovation and stabilitySource
web-artifacts-builderCreate elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn/ui)Source
pypict-claude-skillDesign comprehensive test cases using PICT (Pairwise Independent Combinatorial Testing) for optimized test suitesSource
aws-skillsAWS development with CDK best practices, cost optimization, and serverless/event-driven architecture patternsSource
move-code-quality-skillAnalyzes Move language packages against the official Move Book Code Quality ChecklistSource
audit-websiteCli website auditing tool for seo, performance, security and 140+ other rulesSource
stripe-best-practicesOfficial Stripe skill for building correct integrations with payments, subscriptions, and webhooksSource
upgrade-stripeOfficial Stripe skill for safely upgrading SDK and API versionsSource
expo-app-designOfficial Expo skill for designing and building Expo/React Native apps with best practicesSource
supabase-postgresOfficial Supabase skill for PostgreSQL best practices in Supabase projectsSource
terraform-code-generationOfficial HashiCorp skill for generating and validating Terraform HCL codeSource
terraform-module-generationOfficial HashiCorp skill for creating and refactoring Terraform modulesSource
terraform-provider-developmentOfficial HashiCorp skill for developing custom Terraform providersSource
terraform-skillComprehensive Terraform/OpenTofu best practices covering testing, modules, CI/CD, and 100+ patternsSource
cloudflare-agents-sdkOfficial Cloudflare skill for building stateful AI agents with scheduling, RPC, and MCP on WorkersSource
cloudflare-wranglerOfficial Cloudflare skill for deploying and managing Workers, KV, R2, D1, and QueuesSource
cloudflare-web-perfOfficial Cloudflare skill for auditing Core Web Vitals and render-blocking resourcesSource
cloudflare-building-ai-agentOfficial Cloudflare skill for building full AI agents on Cloudflare's infrastructureSource
cloudflare-building-mcp-serverOfficial Cloudflare skill for building and deploying MCP servers on Cloudflare WorkersSource
cloudflare-durable-objectsOfficial Cloudflare skill for building stateful, globally consistent applications with Durable ObjectsSource
cloudflare-sandbox-sdkOfficial Cloudflare skill for running untrusted code safely using Cloudflare's sandbox environmentSource
cloudflare-workers-best-practicesOfficial Cloudflare skill for Workers best practices covering performance, security, and deployment patternsSource
netlify-functionsOfficial Netlify skill for building serverless API endpoints and background tasksSource
netlify-dbOfficial Netlify skill for managed Postgres with deploy preview branchingSource
neon-postgresOfficial Neon skill for best practices with Neon Serverless PostgresSource
vercel-reactOfficial Vercel skill for React best practices and patternsSource
next-best-practicesOfficial Vercel skill for Next.js recommended patterns and best practicesSource
next-upgradeOfficial Vercel skill for upgrading Next.js projects to newer versionsSource
react-native-best-practicesOfficial Callstack skill for React Native performance optimization patternsSource
better-authOfficial Better Auth skill for authentication integration including OAuth, 2FA, and passkeysSource
tinybirdOfficial Tinybird skill for project guidelines covering datasources, pipes, endpoints, and SQLSource
sanityOfficial Sanity skill for Sanity Studio, GROQ queries, and content workflowsSource
clickhouseOfficial ClickHouse skill for analytics query best practicesSource
remotion-skillOfficial Remotion skill for programmatic video creation with ReactSource
ios-simulator-skill21-script toolkit for iOS simulator automation with accessibility APIs and 96% token reductionSource
claude-d3js-skillInteractive data visualizations using D3.js with reference materials and patternsSource
playwright-skillGeneral-purpose browser automation with Playwright for testing and web interactionSource
claude-a11y-skillComprehensive accessibility audits combining axe-core and eslint-plugin-jsx-a11y against WCAG 2.1Source
context-engineering-kitAdvanced context engineering with multi-agent patterns, reflexion loops, and domain-driven developmentSource
compound-engineering-pluginPragmatic engineering plugin with ideation, planning, execution, multi-agent review, and knowledge compoundingSource
vscode-extension-builderScaffolds, compiles, packages, and installs a working VS Code extension end-to-end from a single request — command extensions, color themes, and snippet packsSource
ponytailForces the laziest solution that actually works — YAGNI, standard library before custom code, one line before fifty, with lite/full/ultra intensity levelsSource
agent-browserOfficial Vercel browser automation CLI for AI agents — navigate pages, fill forms, click, screenshot, extract data, and test web appsSource
prisma-client-apiOfficial Prisma reference for Client API queries, filters, operators, transactions, and client configurationSource
google-agents-cli-deployOfficial Google skill for deploying ADK agents to Agent Runtime, Cloud Run, or GKE, with CI/CD, secrets, and Agent Gateway setupSource
microsoft-foundryOfficial Microsoft skill for building, deploying, evaluating, and managing Foundry agents, models, and resources end to endSource

| prompt-engineer | Turns a request into a production-ready prompt with test cases and success criteria, and routes to a hook, skill, or subagent when a prompt is the wrong artifact | Source | | orca-replay | Answers questions about a past agent run from its recording rather than from memory — what it sent, ran and changed — and replays or forks that run. Reads traces over MCP; needs the orcareplay npm package | Source |

📊 Data & Analysis

NameDescriptionLink
csv-data-summarizerGenerate statistics and charts from CSVsSource
root-cause-tracingTrace and diagnose the source of data or logic errorsSource
postgresExecute safe read-only SQL queries against PostgreSQL databases with multi-connection supportSource
read-only-postgresSafe read-only PostgreSQL queries with multi-connection support and defense-in-depth securitySource
prompt-template-wizardConverts incomplete feature/bug requests into complete structured prompt templatesSource
monte-carlo-agent-toolkitOfficial data and agent observability skills for monitoring, triage, troubleshooting, and health checksSource

🔬 Scientific & Research Tools

NameDescriptionLink
claude-scientific-skills125+ scientific skills for bioinformatics, cheminformatics, clinical research, and machine learningSource
materials-simulation-skillsAgent skills for computational materials science: numerical stability, time-stepping, linear solvers, and simulation validationSource
extract-from-pdfsExtract structured data from scientific PDFs using Claude's vision with external validation against GBIF/WFO databasesSource
phylo-from-buscosPhylogenomic workflows from genome assemblies using BUSCO single-copy orthologsSource
claude-mountaineering-skillsMountain route research aggregating 10+ sources including weather, avalanche forecasts, and hazard analysisSource

📝 Writing & Research

NameDescriptionLink
article-extractorExtract full content from web articlesSource
Content Research WriterResearch and refine written content with feedbackSource
internal-commsDraft formal internal comms and reportsSource
writing-plansCreate structured written plans with clear milestonesSource
writing-skillsEnhance instructional and technical writing qualitySource
brainstormingFacilitate creative idea generation sessionsSource
family-history-researchProvides assistance with planning family history and genealogy research projectsSource
avoid-ai-writingAudit and rewrite content to remove 21 categories of AI writing patterns with a 43-entry replacement tableSource
non-fiction-book-factoryComplete nonfiction book creation pipeline replicating traditional publishing infrastructureSource
ebook-factoryFocused ebook creation pipeline for producing publication-ready digital booksSource
academic-research-skills13-agent pipeline for rigorous academic research with Socratic guided mode and systematic review/PRISMA supportSource
academic-paper12-agent paper writing system with LaTeX output, visualization generation, revision coaching, and multi-format citation conversionSource
academic-paper-reviewerMulti-perspective peer review with 0-100 quality rubrics, Editor-in-Chief, 3 dynamic reviewers, and Devil's Advocate rolesSource
claude-scientific-writerPublication-ready scientific document generation with real-time research lookup, citation management, and 8-dimension ScholarEval reviewSource
claude-blogBlog content creation ecosystem with 12 commands for writing, rewriting, SEO, schema markup, content repurposing, and editorial calendarsSource
claude-emailAI-powered email management with inbox triage, composition using copywriting frameworks (PAS, AIDA, BAB), deliverability audits, and automation sequencesSource
autoresearchAutonomous research orchestration using a two-loop architecture managing the full lifecycle from literature survey to paper writingSource
writing-fragmentsMine raw fragments through relentless grilling — pure explore mode with no structure imposedSource
writing-beatsAssemble raw material into a journey of beats, grounding each term before a beat leans on itSource
writing-shapeShape raw material into an article, paragraph by paragraph, from a pile of notes or fragmentsSource
humanizerRewrites AI-sounding text so it reads like a person wrote it, without changing what it says — based on Wikipedia's "Signs of AI writing"Source
no-ai-slopEdits drafts into sharper, more human writing while preserving the writer's personal voice, removing 20+ patterns of AI slopSource

📚 Learning & Knowledge

NameDescriptionLink
tapestryBuild a linked knowledge graph from documentsSource
ship-learn-nextRecommend next steps based on feedback loopsSource
using-superpowersLearn and apply best practices for Superpowers workflowsSource
sharing-skillsLearn how to contribute new skills via pull requestsSource
collision-zone-thinkingCombine unrelated concepts to find new creative or problem-solving connectionsSource
inversion-exerciseFlip assumptions to uncover hidden insights and constraintsSource
meta-pattern-recognitionIdentify patterns across domains to uncover universal principlesSource
scale-gameStress-test ideas at extreme scales to expose hidden weaknesses or truthsSource
simplification-cascadesReduce complexity by discovering insights that simplify multiple elements at onceSource
tracing-knowledge-lineagesTrack how ideas evolve across iterations and influencesSource
think-deeplyPrevents confirmatory answers by encouraging nuanced multi-perspective analysis for complex questionsSource
obsidian-markdownStructured markdown authoring and knowledge workflows optimized for Obsidian vaultsSource
json-canvasWork with Obsidian JSON Canvas for visual knowledge mapping and linkingSource
teachLearn a concept over multiple sessions using the current directory as a stateful workspace with glossary, missions, and resourcesSource

| teaching-skills | University teaching-lifecycle suite for professors: course design, lessons, assessment, grading, and reflection (15 skills) | Source |

🎥 Media & Content

NameDescriptionLink
youtube-transcriptSummarize YouTube transcriptsSource
claude-epub-skillParse and analyze EPUB eBooksSource
Image EnhancerImprove resolution and clarity of screenshotsSource
Video DownloaderDownload YouTube videos for use in ClaudeSource
imagenGenerate images using Google Gemini's image generation API for UI mockups, icons, and visual assetsSource
typefullyOfficial Typefully skill for creating and scheduling social media content across X, LinkedIn, Threads, Bluesky, and MastodonSource
video-editSmart router that matches video-editing intent to the right model in the RunComfy catalog — restyle, background swap, and packaging with identity and motion preservationSource
hyperframesHeyGen's agent-native HTML-to-video framework: CLI development loop with keyframes, capture, render, composition, and publishing skillsSource

| ruxi-skill | Turn any book into a playable visual novel where you play an overlooked side character, every branch cited back to the source text, output as a single playable HTML file | Source | | socialclaw | Social media scheduling and publishing for AI agents across X, LinkedIn, Instagram, Facebook Pages, TikTok, Discord, Telegram, YouTube, Reddit, WordPress, and Pinterest via one workspace API key | Source |

🤝 Collaboration & Project Management

NameDescriptionLink
agent-workflow-packPackage repeatable AI coding workflows into SKILL.md, templates, and verification checksSource
grill-meStructured questioning workflow to pressure-test plans and designs before executionSource
handoffGenerate concise handoff documentation so another agent can continue work seamlesslySource
grillingGrill the user relentlessly about a plan or design — one question at a time with a recommended answerSource
loop-meGrill about specs for the workflows you want to build, using the current directory as a stateful workspaceSource
Meeting Insights AnalyzerAnalyze meeting dynamics and communication patternsSource
Notion Integration SkillsOfficial Notion connectors for ClaudeSource
commandsManage and automate recurring project commandsSource
receiving-code-reviewProcess and apply code review feedbackSource
requesting-code-reviewRequest and manage structured code reviewsSource
dispatching-parallel-agentsCoordinate multiple Claude subagents on shared tasksSource
remembering-conversationsRecall facts, insights, and context from past Claude Code sessionsSource
git-pushingAutomate git operations and repository interactionsSource
linear-claude-skillManage Linear issues, projects, and teams with MCP tools and GraphQL fallbacksSource
linear-cli-skillA skill teaching Claude how to use linear-CLI as an alternative to Linear MCPSource
review-implementingEvaluate code implementation plans and align with specsSource
test-fixingDetect failing tests and propose patches or fixesSource
product-manager-skillsSenior PM agent with 6 knowledge domains, 12 templates, 30+ frameworks covering discovery, strategy, delivery, SaaS metrics, PM career coaching, and AI product craftSource
read-only-gh-pr-reviewSafe read-only review of backend pull requests using GitHub CLI with local inspectionSource
i-have-adhdShapes agent output for ADHD readers: lead with the next action, number multi-step work, restate state across turns, and suppress tangentsSource

| yylo-skills | Seven reusable agent skills for Claude Code, Codex, and Pi built around YYLO Ledger — operate task management, maintain wiki and workflow Records, capture provenance-bound evidence, plan implementation-sized tasks, execute one assigned task through validated delivery, and inspect the project before implementation | Source |

📣 Marketing & SEO

NameDescriptionLink
copywritingExpert conversion copywriter skill that prioritizes clarity and business outcomes with proven frameworksSource
seo-auditTechnical and on-page SEO review for websitesSource
ai-seoAI search engine optimization for visibility in AI-powered search resultsSource
programmatic-seoScaled SEO page generation for large content operationsSource
content-strategyContent planning and topic research for marketing teamsSource
copy-editingMarketing copy review and refinement for quality and conversionSource
email-sequenceAutomated email flow development for welcome, nurture, and re-engagement seriesSource
cold-emailB2B outreach email sequence creation and optimizationSource
ad-creativeAd headline and creative generation for paid campaignsSource
ab-test-setupA/B testing and experiment planning for marketing campaignsSource
page-croMarketing page conversion rate optimizationSource
pricing-strategyPricing and monetization strategy for SaaS and productsSource
launch-strategyProduct launch planning and go-to-market strategySource
social-contentSocial media content creation across platformsSource
lead-magnetsLead magnet creation and optimization for list buildingSource
referral-programReferral and affiliate program design and implementationSource
sales-enablementSales collateral and pitch deck creation for B2B teamsSource
analytics-trackingAnalytics setup and measurement implementationSource
marketing-psychologyBehavioral science application to marketing copy and UXSource
seo-audit-fullDeep technical and content SEO auditing workflow for comprehensive site/page analysisSource
vibe-prospectingLive B2B company and contact data for natural-language lead-list building, prospect enrichment, executive discovery, and multi-step GTM workflowsSource
reddit-automationFind Reddit threads where people are genuinely asking for what you offer, then draft honest, helpful replies with affiliation disclosureSource

NameDescriptionLink
proficiently-claude-skillsAI-powered job search, resume tailoring, and cover letter writingSource
claude-code-job-tailorResume optimization system that analyzes job postings, ranks requirements, and generates tailored PDFs in under 60 secondsSource
resume-tailoring-skillAI-powered resume tailoring optimized for specific job descriptions while maintaining factual integritySource
claude-resume-kitExtract experience once, generate tailored LaTeX resumes per JD with anti-fabrication controls and multi-perspective critiqueSource
Resume-BuilderAI-powered resume and cover letter generator with dual ATS + HR scoring for any professionSource
ResumeSkillsResume optimization, ATS scoring, interview prep, and strategic job search skills collectionSource

🔐 Security & Testing

NameDescriptionLink
Trail of Bits Security SkillsSecurity skills for static analysis with CodeQL/Semgrep, variant analysis, code auditing, and fix verificationSource
webapp-testingUI test automation using PlaywrightSource
ffuf_claude_skillFuzz test web apps with FFUF + ClaudeSource
defense-in-depthImplement multi-layered testing and security best practicesSource
systematic-debuggingStructured debugging with hypothesis testing and validationSource
testing-anti-patternsIdentify and prevent testing anti-patternsSource
testing-skills-with-subagentsVerify new skills using subagents and test cyclesSource
verification-before-completionRun verification checks before closing tasksSource
condition-based-waitingUse logical conditions to control test flow timingSource
varlock-claude-skillSecure environment variable management ensuring secrets never appear in sessions, terminals, logs, or git commitsSource
security-fuzzingEssential fuzzing payloads for SQL injection, command injection, NoSQL injection, and LDAP injection testingSource
security-payloadsSpecialized attack payloads for XSS vectors, XXE, template injection, and file upload bypass testingSource
llm-testingAI/ML security testing including bias detection, data leakage, alignment testing, and adversarial resistanceSource
accessibility-skillsAccessibility guidance for accessible interfaces, including forms, keyboard use, color contrast, and light/dark modesSource
security-hardeningPortable defensive security workflow covering secure design and review, OWASP, MCP security, prompt injection, incident response, and hardeningSource

⚙️ Utility & Automation

NameDescriptionLink
writing-great-skillsReference for writing and editing skills well — the vocabulary and principles that make a skill predictableSource
git-guardrails-claude-codeInstall guardrails to block dangerous git commands before executionSource
setup-pre-commitConfigure Husky pre-commit hooks with lint-staged, formatting, and quality checksSource
file-organizerClean up file structures, rename documentsSource
invoice-organizerParse and categorize invoicesSource
raffle-winner-pickerPick winners using secure randomnessSource
skill-creatorBuild your own skill interactivelySource
template-skillA starting template for new skillsSource
gardening-skills-wikiMaintain the skills wiki, ensuring naming consistency and metadata qualitySource
pulling-updates-from-skills-repositorySync and pull the latest skill updates from repositoriesSource
cc-devops-skills31 DevOps skills covering IaC, CI/CD, Kubernetes, observability, and scripting automationSource
opentelemetry-skillsInstrumentation, telemetry-quality, collector, semantic-convention, and OTTL guidance for observable applicationsSource
devops-claude-skillsDevOps workflow skills for Terraform, K8s troubleshooting, AWS cost optimization, and GitOpsSource
aws-well-architected-reviewReusable playbooks for applying the AWS Well-Architected Framework to architecture reviews and improvement plansSource
firecrawl-cliOfficial Firecrawl skill for scraping, crawling, searching, and mapping the web via CLISource
hermes-tweetNative Hermes Agent plugin skill for X monitoring, account research, trend checks, and approval-gated social actions. Xquik is an independent third-party service. Not affiliated with X Corp. "Twitter" and "X" are trademarks of X Corp.Source
replicateOfficial Replicate skill for discovering, comparing, and running AI models via APISource
agentsys36 workflow automation skills for profiling, code review, AI consultation, release management, and drift analysisSource
better-i18nOfficial i18n skill for internationalization best practices, translation workflows, and localization automationSource
migrate-to-shoehornMigrate test assertions to @total-typescript/shoehorn patternsSource
scaffold-exercisesScaffold structured exercise directories with sections, problems, solutions, and explainersSource
edit-articleAssist with article editing and refinement workflowsSource
obsidian-vaultWork with and organize notes in an Obsidian vaultSource
obsidian-cliUse Obsidian CLI workflows to automate vault operations and content managementSource
wizardGenerate an interactive bash wizard that walks a human through a manual procedure — opening URLs, capturing values, confirming each stepSource
setup-ts-deep-modulesWire dependency-cruiser into a TypeScript repo so each package is a deep module with a small public surfaceSource
claude-handoffHand the current conversation off to a fresh background agent that picks up the work immediatelySource
find-skillsOfficial Vercel skill that discovers and installs agent skills from the ecosystem when you ask "is there a skill for X"Source

| danshari-skill (断舍离.skill) | Audits every skill you have installed and archives the ones your current model, harness, or MCP servers already cover — three-way triage, blind-test evidence before any removal, and one-command restore instead of rm | Source | | sijiao-skill (私教.skill) | Distils any skill you want to learn into a stateful tutor — it remembers where you are, sets and grades exercises at your level, and schedules spaced review, with the ceiling honestly capped at competent | Source |

Getting Started

Install across supported agents

Skills use the portable SKILL.md format. The open-source Skills CLI installs compatible skills for 70+ coding agents, including Claude Code, Codex, Z.ai's ZCode, Cursor, GitHub Copilot, Gemini CLI, and OpenCode. Copy a source link from a row above, then list and install only the skill you want:

# Inspect the skill names available from a source repository.
npx skills add https://github.com/anthropics/skills --list

# Try a skill without installing it (replace the agent ID as needed).
npx skills use https://github.com/anthropics/skills --skill claude-api --agent claude-code

# Install a specific skill for the current project (the default scope).
npx skills add https://github.com/anthropics/skills/tree/main/skills/claude-api --agent claude-code

# Make it available in every project, or install it for another supported agent.
npx skills add https://github.com/anthropics/skills/tree/main/skills/claude-api --agent codex --global
ToolSkills CLI agent IDProject pathGlobal path
Claude Codeclaude-code.claude/skills/~/.claude/skills/
Codexcodex.agents/skills/~/.codex/skills/
Z.ai ZCodezcode.zcode/skills/~/.zcode/skills/
Cursorcursor.agents/skills/~/.cursor/skills/
GitHub Copilotgithub-copilot.agents/skills/~/.copilot/skills/
Gemini CLIgemini-cli.agents/skills/~/.gemini/skills/

For ZCode, you can also use Settings → Skills → Import to scan and import skills already installed for Claude Code, Codex, and other supported agents. For a tool not listed here, place the skill directory containing SKILL.md in that tool's documented project or user skills folder, then refresh/restart the tool.

Use the interactive prompts to choose the installation method and review the files. Do not use --all or -y unless you have inspected and trust every selected skill. To check installed skills or refresh them later:

npx skills list
npx skills update

Let an agent tailor the selection

Paste this into Claude Code, Codex, ZCode, or another coding agent from inside your project:

Audit this repository read-only and recommend a small, personalized set of skills from https://github.com/abubakarsiddik31/claude-skills-collection. Match the recommendations to the stack, workflows, and current pain points. Exclude NSFW content and programming-language-specific skills. For each recommendation, show the source URL, the exact skill name, prerequisites, data/credential implications, and why it fits. Inspect the skill files before recommending installation. Do not install, change configuration, or use non-interactive flags until I explicitly approve the exact list.

After you approve the list, ask the agent to install the selected skills at project scope for shared team workflows or global scope for a personal toolkit. Name your host explicitly—for example claude-code, codex, or zcode—and pin a source revision when reproducibility matters.


Skill Quality Standards & Template

Writing a great skill starts with a well-structured SKILL.md file. AgentSkills.io provides the canonical specification for the Claude Skills format, including validation rules and a detailed checklist. Anthropic also publishes best practices for skill authoring at platform.claude.com.

Minimal SKILL.md Template

Every skill should include a SKILL.md with proper YAML frontmatter at the top:

---
name: my-skill-name
description: |
  Generates weekly status reports from project data.
  Use when the user asks for a summary of recent activity.
---

The body of SKILL.md (below the frontmatter) contains the instructions Claude follows when the skill is activated.

Key Quality Criteria

  • name -- Use kebab-case, 1-64 characters (e.g., csv-data-summarizer, not CSV Data Summarizer).
  • description -- Write in third person. State what the skill does and when to use it. Keep it under 300 characters.
  • Body length -- Keep the instruction body under 500 lines. Long skills should be split into focused sub-skills.
  • Progressive disclosure -- Start with the most common usage, then cover edge cases. Claude reads top-down, so put the critical path first.
  • No secrets in source -- Never embed API keys, tokens, or credentials in SKILL.md. Use environment variables or MCP configuration instead.

Reference Implementations


Contributing

Have a skill to add?
Open a pull request or submit your repo link in an issue with:

  • Name of the skill
  • Short description
  • Category
  • Link to the source

License

This repo lists and links to skills under various licenses.
Please refer to each linked repository for license terms.


Significant stargazers

Trevin Chow

571 followers · starred Oct 2025

Agustin Rivera

41 followers · starred Dec 2025

abubakarsiddik31/claude-skills-collection

A curated collection of official and community-built Claude Skills – extend Anthropic's Claude with powerful, modular capabilities for productivity, creativity, coding, and more.

1,086

61 commits

updated Sep 12, 2026

See the code

README

📦 Claude Skills Collection

A curated collection of official and community-built Claude Skills.
Anthropic Skills are modular tools that extend Claude with practical workflows for coding, debugging, planning, writing, research, design, and automation.

This repository gathers and organizes all publicly available Claude Skills, including both built-in tools by Anthropic and creative contributions from the community. Browse by category, explore capabilities, and kickstart your own Skill creation.

💡 Note: Skills require Claude Pro, Max, Team, or Enterprise access with code execution enabled.

If you use skills across more than one coding agent, Skill Manager gives you one dashboard to discover, enable, disable, edit, and delete them across Claude Code, Codex, Cursor, and OpenCode. Download it here; view the source on GitHub.

Skill Manager — every skill, every agent, one dashboard

📈 Overview

252 skills across 13 categories:

CategorySkills
💻 Development & Code Tools82
📣 Marketing & SEO22
📝 Writing & Research22
🤝 Collaboration & Project Management21
⚙️ Utility & Automation30
🔐 Security & Testing15
📚 Learning & Knowledge15
🎨 Creative & Design13
💼 Career & Job Search6
🎥 Media & Content10
🔬 Scientific & Research Tools5
📊 Data & Analysis6
📄 Document Skills5

These are commonly requested, high-impact skills to start with:

  • grill-me - pressure-test plans and assumptions before execution
  • grill-with-docs - structured planning with CONTEXT.md and ADR updates
  • tdd - red/green/refactor workflow for safer implementation
  • diagnosing-bugs - disciplined root-cause debugging loop
  • handoff - compact transfer notes between sessions/agents

📚 Table of Contents


What Are Claude Skills?

Claude Skills are specialized modules that Claude can use to perform structured, multi-step tasks. Each skill is a lightweight folder with a clear instruction interface, and optionally includes Python code, templates, and assets.

Skills allow Claude to:

  • Create and edit documents (Word, Excel, PPT)
  • Parse and analyze structured data
  • Write and debug code
  • Generate creative visual assets
  • Automate research, testing, or collaboration tasks

Official Skills are created by Anthropic and auto-invoked when needed. You can also install or build custom skills to meet your unique workflow needs.


📄 Document Skills

NameDescriptionLink
docxCreate and edit Microsoft Word documents with formatting, comments, and tracked changesSource
pdfExtract content from PDFs, split/merge documents, or create new onesSource
pptxGenerate and edit PowerPoint presentationsSource
xlsxManipulate Excel files, formulas, tables, and chartsSource
revealjs-skillGenerate polished, professional presentations using the Reveal.js HTML presentation frameworkSource

🎨 Creative & Design

NameDescriptionLink
algorithmic-artCreate generative art using p5.jsSource
canvas-designRender layout-based visual designs in PNG/PDFSource
slack-gif-creatorGenerate Slack-optimized animated GIFsSource
brand-guidelinesApply company branding to outputsSource
theme-factoryCreate and apply visual themes for documentsSource
nano-banana-image-generationCreate images using Nano Banana ProSource
frontend-slidesCreate animation-rich single-file HTML presentations with 12 visual styles and zero dependenciesSource
web-asset-generatorGenerate favicons, app icons, PWA manifests, and Open Graph images with WCAG contrast validationSource
color-expertColor science and accessibility expertise covering APCA/WCAG, palette systems, and color-space decisionsSource
design-mobile-appsDesign mobile app UI screens and implement Sleek designs in HTML, React Native, or SwiftUISource
frontend-designOfficial Anthropic guidance for distinctive, intentional visual design — aesthetic direction, typography, and avoiding templated defaultsSource
web-design-guidelinesOfficial Vercel skill that reviews UI code against Web Interface Guidelines for accessibility, interaction, and UX best practicesSource
emil-design-engEmil Kowalski's design-engineering philosophy: UI polish, component design, animation decisions, and the invisible details that make software feel greatSource

💻 Development & Code Tools

NameDescriptionLink
code-reviewTwo-axis review of changes since a fixed point — Standards (coding standards) and Spec (originating issue/PRD) — run in parallel sub-agentsSource
claude-apiOfficial Anthropic guidance for building, debugging, and optimizing Claude API applications, including tool use, prompt caching, batches, files, citations, and memorySource
claude-skills-pro15 engineering-workflow skills (7-dimension code review, root-cause bug hunting, test generation with mutation check, zero-downtime DB migrations, security audit) — 7 free under MIT, CN handbook includedSource
codebase-designShared vocabulary for designing deep modules — lots of behaviour behind a small interface at a clean seamSource
diagnosing-bugsDisciplined diagnosis loop for hard bugs and performance regressions: reproduce, minimize, hypothesize, instrument, fix, and regression-testSource
domain-modelingBuild and sharpen a project's domain model with ubiquitous language, glossary entries, and architectural decision recordsSource
grill-with-docsGrilling workflow that pressure-tests plans against domain language and updates CONTEXT.md and ADRsSource
implementImplement a piece of work based on a spec or set of tickets, driving TDD internally per issueSource
improve-codebase-architectureFind deepening opportunities in a codebase using shared domain language and architectural decisionsSource
prototypeBuild throwaway prototypes for logic or UI exploration before production implementationSource
researchDelegate reading legwork to a background agent that investigates against primary sources and leaves a cited Markdown fileSource
resolving-merge-conflictsResolve in-progress git merge/rebase conflicts by understanding both changes' intent and preserving itSource
setup-matt-pocock-skillsOne-time repository setup for issue tracker, triage labels, and docs layout used by the engineering skill suiteSource
tddTest-driven development workflow using a red-green-refactor loop and vertical slicesSource
to-specTurn the current conversation into a spec and publish it to the project issue tracker — synthesis, no interviewSource
to-ticketsBreak a plan, spec, or conversation into tracer-bullet tickets, each declaring its blocking edgesSource
triageTriage work through a structured issue-state workflow with explicit rolesSource
wayfinderPlan huge multi-session work as a shared map of investigation tickets, resolving them one at a time until the route is clearSource
interview-meInteractive requirements interview loop to uncover hidden assumptions before implementationSource
source-driven-developmentGround implementation in primary sources and existing code reality before writing changesSource
performance-optimizationDiagnose and improve runtime, render, and bundle performance with systematic profilingSource
code-review-and-qualityEnforce production-grade review criteria, maintainability, and quality gatesSource
incremental-implementationShip complex features as small validated slices to reduce risk and regressionsSource
planning-and-task-breakdownConvert broad goals into sequenced, executable engineering tasksSource
swiftui-proAdvanced SwiftUI development patterns and implementation guidance for Apple platformsSource
swift-concurrency-proPractical Swift Concurrency patterns for structured async code and correctnessSource
react-components-stitchGenerate robust React components using Stitch-oriented implementation patternsSource
MCP ServerBuild Claude-compatible API connectorsSource
Changelog GeneratorCreate changelogs from commit historySource
using-git-worktreesManage feature branches safely in isolated Git worktreesSource
test-driven-developmentWrite tests before implementation to drive developmentSource
subagent-driven-developmentUse multiple Claude subagents to coordinate complex implementationsSource
executing-plansExecute structured plans with checkpoints and verification stepsSource
finishing-a-development-branchComplete development branches with testing and review flowSource
preserving-productive-tensionsManage architectural decisions by preserving competing viewpoints to balance innovation and stabilitySource
web-artifacts-builderCreate elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn/ui)Source
pypict-claude-skillDesign comprehensive test cases using PICT (Pairwise Independent Combinatorial Testing) for optimized test suitesSource
aws-skillsAWS development with CDK best practices, cost optimization, and serverless/event-driven architecture patternsSource
move-code-quality-skillAnalyzes Move language packages against the official Move Book Code Quality ChecklistSource
audit-websiteCli website auditing tool for seo, performance, security and 140+ other rulesSource
stripe-best-practicesOfficial Stripe skill for building correct integrations with payments, subscriptions, and webhooksSource
upgrade-stripeOfficial Stripe skill for safely upgrading SDK and API versionsSource
expo-app-designOfficial Expo skill for designing and building Expo/React Native apps with best practicesSource
supabase-postgresOfficial Supabase skill for PostgreSQL best practices in Supabase projectsSource
terraform-code-generationOfficial HashiCorp skill for generating and validating Terraform HCL codeSource
terraform-module-generationOfficial HashiCorp skill for creating and refactoring Terraform modulesSource
terraform-provider-developmentOfficial HashiCorp skill for developing custom Terraform providersSource
terraform-skillComprehensive Terraform/OpenTofu best practices covering testing, modules, CI/CD, and 100+ patternsSource
cloudflare-agents-sdkOfficial Cloudflare skill for building stateful AI agents with scheduling, RPC, and MCP on WorkersSource
cloudflare-wranglerOfficial Cloudflare skill for deploying and managing Workers, KV, R2, D1, and QueuesSource
cloudflare-web-perfOfficial Cloudflare skill for auditing Core Web Vitals and render-blocking resourcesSource
cloudflare-building-ai-agentOfficial Cloudflare skill for building full AI agents on Cloudflare's infrastructureSource
cloudflare-building-mcp-serverOfficial Cloudflare skill for building and deploying MCP servers on Cloudflare WorkersSource
cloudflare-durable-objectsOfficial Cloudflare skill for building stateful, globally consistent applications with Durable ObjectsSource
cloudflare-sandbox-sdkOfficial Cloudflare skill for running untrusted code safely using Cloudflare's sandbox environmentSource
cloudflare-workers-best-practicesOfficial Cloudflare skill for Workers best practices covering performance, security, and deployment patternsSource
netlify-functionsOfficial Netlify skill for building serverless API endpoints and background tasksSource
netlify-dbOfficial Netlify skill for managed Postgres with deploy preview branchingSource
neon-postgresOfficial Neon skill for best practices with Neon Serverless PostgresSource
vercel-reactOfficial Vercel skill for React best practices and patternsSource
next-best-practicesOfficial Vercel skill for Next.js recommended patterns and best practicesSource
next-upgradeOfficial Vercel skill for upgrading Next.js projects to newer versionsSource
react-native-best-practicesOfficial Callstack skill for React Native performance optimization patternsSource
better-authOfficial Better Auth skill for authentication integration including OAuth, 2FA, and passkeysSource
tinybirdOfficial Tinybird skill for project guidelines covering datasources, pipes, endpoints, and SQLSource
sanityOfficial Sanity skill for Sanity Studio, GROQ queries, and content workflowsSource
clickhouseOfficial ClickHouse skill for analytics query best practicesSource
remotion-skillOfficial Remotion skill for programmatic video creation with ReactSource
ios-simulator-skill21-script toolkit for iOS simulator automation with accessibility APIs and 96% token reductionSource
claude-d3js-skillInteractive data visualizations using D3.js with reference materials and patternsSource
playwright-skillGeneral-purpose browser automation with Playwright for testing and web interactionSource
claude-a11y-skillComprehensive accessibility audits combining axe-core and eslint-plugin-jsx-a11y against WCAG 2.1Source
context-engineering-kitAdvanced context engineering with multi-agent patterns, reflexion loops, and domain-driven developmentSource
compound-engineering-pluginPragmatic engineering plugin with ideation, planning, execution, multi-agent review, and knowledge compoundingSource
vscode-extension-builderScaffolds, compiles, packages, and installs a working VS Code extension end-to-end from a single request — command extensions, color themes, and snippet packsSource
ponytailForces the laziest solution that actually works — YAGNI, standard library before custom code, one line before fifty, with lite/full/ultra intensity levelsSource
agent-browserOfficial Vercel browser automation CLI for AI agents — navigate pages, fill forms, click, screenshot, extract data, and test web appsSource
prisma-client-apiOfficial Prisma reference for Client API queries, filters, operators, transactions, and client configurationSource
google-agents-cli-deployOfficial Google skill for deploying ADK agents to Agent Runtime, Cloud Run, or GKE, with CI/CD, secrets, and Agent Gateway setupSource
microsoft-foundryOfficial Microsoft skill for building, deploying, evaluating, and managing Foundry agents, models, and resources end to endSource

| prompt-engineer | Turns a request into a production-ready prompt with test cases and success criteria, and routes to a hook, skill, or subagent when a prompt is the wrong artifact | Source | | orca-replay | Answers questions about a past agent run from its recording rather than from memory — what it sent, ran and changed — and replays or forks that run. Reads traces over MCP; needs the orcareplay npm package | Source |

📊 Data & Analysis

NameDescriptionLink
csv-data-summarizerGenerate statistics and charts from CSVsSource
root-cause-tracingTrace and diagnose the source of data or logic errorsSource
postgresExecute safe read-only SQL queries against PostgreSQL databases with multi-connection supportSource
read-only-postgresSafe read-only PostgreSQL queries with multi-connection support and defense-in-depth securitySource
prompt-template-wizardConverts incomplete feature/bug requests into complete structured prompt templatesSource
monte-carlo-agent-toolkitOfficial data and agent observability skills for monitoring, triage, troubleshooting, and health checksSource

🔬 Scientific & Research Tools

NameDescriptionLink
claude-scientific-skills125+ scientific skills for bioinformatics, cheminformatics, clinical research, and machine learningSource
materials-simulation-skillsAgent skills for computational materials science: numerical stability, time-stepping, linear solvers, and simulation validationSource
extract-from-pdfsExtract structured data from scientific PDFs using Claude's vision with external validation against GBIF/WFO databasesSource
phylo-from-buscosPhylogenomic workflows from genome assemblies using BUSCO single-copy orthologsSource
claude-mountaineering-skillsMountain route research aggregating 10+ sources including weather, avalanche forecasts, and hazard analysisSource

📝 Writing & Research

NameDescriptionLink
article-extractorExtract full content from web articlesSource
Content Research WriterResearch and refine written content with feedbackSource
internal-commsDraft formal internal comms and reportsSource
writing-plansCreate structured written plans with clear milestonesSource
writing-skillsEnhance instructional and technical writing qualitySource
brainstormingFacilitate creative idea generation sessionsSource
family-history-researchProvides assistance with planning family history and genealogy research projectsSource
avoid-ai-writingAudit and rewrite content to remove 21 categories of AI writing patterns with a 43-entry replacement tableSource
non-fiction-book-factoryComplete nonfiction book creation pipeline replicating traditional publishing infrastructureSource
ebook-factoryFocused ebook creation pipeline for producing publication-ready digital booksSource
academic-research-skills13-agent pipeline for rigorous academic research with Socratic guided mode and systematic review/PRISMA supportSource
academic-paper12-agent paper writing system with LaTeX output, visualization generation, revision coaching, and multi-format citation conversionSource
academic-paper-reviewerMulti-perspective peer review with 0-100 quality rubrics, Editor-in-Chief, 3 dynamic reviewers, and Devil's Advocate rolesSource
claude-scientific-writerPublication-ready scientific document generation with real-time research lookup, citation management, and 8-dimension ScholarEval reviewSource
claude-blogBlog content creation ecosystem with 12 commands for writing, rewriting, SEO, schema markup, content repurposing, and editorial calendarsSource
claude-emailAI-powered email management with inbox triage, composition using copywriting frameworks (PAS, AIDA, BAB), deliverability audits, and automation sequencesSource
autoresearchAutonomous research orchestration using a two-loop architecture managing the full lifecycle from literature survey to paper writingSource
writing-fragmentsMine raw fragments through relentless grilling — pure explore mode with no structure imposedSource
writing-beatsAssemble raw material into a journey of beats, grounding each term before a beat leans on itSource
writing-shapeShape raw material into an article, paragraph by paragraph, from a pile of notes or fragmentsSource
humanizerRewrites AI-sounding text so it reads like a person wrote it, without changing what it says — based on Wikipedia's "Signs of AI writing"Source
no-ai-slopEdits drafts into sharper, more human writing while preserving the writer's personal voice, removing 20+ patterns of AI slopSource

📚 Learning & Knowledge

NameDescriptionLink
tapestryBuild a linked knowledge graph from documentsSource
ship-learn-nextRecommend next steps based on feedback loopsSource
using-superpowersLearn and apply best practices for Superpowers workflowsSource
sharing-skillsLearn how to contribute new skills via pull requestsSource
collision-zone-thinkingCombine unrelated concepts to find new creative or problem-solving connectionsSource
inversion-exerciseFlip assumptions to uncover hidden insights and constraintsSource
meta-pattern-recognitionIdentify patterns across domains to uncover universal principlesSource
scale-gameStress-test ideas at extreme scales to expose hidden weaknesses or truthsSource
simplification-cascadesReduce complexity by discovering insights that simplify multiple elements at onceSource
tracing-knowledge-lineagesTrack how ideas evolve across iterations and influencesSource
think-deeplyPrevents confirmatory answers by encouraging nuanced multi-perspective analysis for complex questionsSource
obsidian-markdownStructured markdown authoring and knowledge workflows optimized for Obsidian vaultsSource
json-canvasWork with Obsidian JSON Canvas for visual knowledge mapping and linkingSource
teachLearn a concept over multiple sessions using the current directory as a stateful workspace with glossary, missions, and resourcesSource

| teaching-skills | University teaching-lifecycle suite for professors: course design, lessons, assessment, grading, and reflection (15 skills) | Source |

🎥 Media & Content

NameDescriptionLink
youtube-transcriptSummarize YouTube transcriptsSource
claude-epub-skillParse and analyze EPUB eBooksSource
Image EnhancerImprove resolution and clarity of screenshotsSource
Video DownloaderDownload YouTube videos for use in ClaudeSource
imagenGenerate images using Google Gemini's image generation API for UI mockups, icons, and visual assetsSource
typefullyOfficial Typefully skill for creating and scheduling social media content across X, LinkedIn, Threads, Bluesky, and MastodonSource
video-editSmart router that matches video-editing intent to the right model in the RunComfy catalog — restyle, background swap, and packaging with identity and motion preservationSource
hyperframesHeyGen's agent-native HTML-to-video framework: CLI development loop with keyframes, capture, render, composition, and publishing skillsSource

| ruxi-skill | Turn any book into a playable visual novel where you play an overlooked side character, every branch cited back to the source text, output as a single playable HTML file | Source | | socialclaw | Social media scheduling and publishing for AI agents across X, LinkedIn, Instagram, Facebook Pages, TikTok, Discord, Telegram, YouTube, Reddit, WordPress, and Pinterest via one workspace API key | Source |

🤝 Collaboration & Project Management

NameDescriptionLink
agent-workflow-packPackage repeatable AI coding workflows into SKILL.md, templates, and verification checksSource
grill-meStructured questioning workflow to pressure-test plans and designs before executionSource
handoffGenerate concise handoff documentation so another agent can continue work seamlesslySource
grillingGrill the user relentlessly about a plan or design — one question at a time with a recommended answerSource
loop-meGrill about specs for the workflows you want to build, using the current directory as a stateful workspaceSource
Meeting Insights AnalyzerAnalyze meeting dynamics and communication patternsSource
Notion Integration SkillsOfficial Notion connectors for ClaudeSource
commandsManage and automate recurring project commandsSource
receiving-code-reviewProcess and apply code review feedbackSource
requesting-code-reviewRequest and manage structured code reviewsSource
dispatching-parallel-agentsCoordinate multiple Claude subagents on shared tasksSource
remembering-conversationsRecall facts, insights, and context from past Claude Code sessionsSource
git-pushingAutomate git operations and repository interactionsSource
linear-claude-skillManage Linear issues, projects, and teams with MCP tools and GraphQL fallbacksSource
linear-cli-skillA skill teaching Claude how to use linear-CLI as an alternative to Linear MCPSource
review-implementingEvaluate code implementation plans and align with specsSource
test-fixingDetect failing tests and propose patches or fixesSource
product-manager-skillsSenior PM agent with 6 knowledge domains, 12 templates, 30+ frameworks covering discovery, strategy, delivery, SaaS metrics, PM career coaching, and AI product craftSource
read-only-gh-pr-reviewSafe read-only review of backend pull requests using GitHub CLI with local inspectionSource
i-have-adhdShapes agent output for ADHD readers: lead with the next action, number multi-step work, restate state across turns, and suppress tangentsSource

| yylo-skills | Seven reusable agent skills for Claude Code, Codex, and Pi built around YYLO Ledger — operate task management, maintain wiki and workflow Records, capture provenance-bound evidence, plan implementation-sized tasks, execute one assigned task through validated delivery, and inspect the project before implementation | Source |

📣 Marketing & SEO

NameDescriptionLink
copywritingExpert conversion copywriter skill that prioritizes clarity and business outcomes with proven frameworksSource
seo-auditTechnical and on-page SEO review for websitesSource
ai-seoAI search engine optimization for visibility in AI-powered search resultsSource
programmatic-seoScaled SEO page generation for large content operationsSource
content-strategyContent planning and topic research for marketing teamsSource
copy-editingMarketing copy review and refinement for quality and conversionSource
email-sequenceAutomated email flow development for welcome, nurture, and re-engagement seriesSource
cold-emailB2B outreach email sequence creation and optimizationSource
ad-creativeAd headline and creative generation for paid campaignsSource
ab-test-setupA/B testing and experiment planning for marketing campaignsSource
page-croMarketing page conversion rate optimizationSource
pricing-strategyPricing and monetization strategy for SaaS and productsSource
launch-strategyProduct launch planning and go-to-market strategySource
social-contentSocial media content creation across platformsSource
lead-magnetsLead magnet creation and optimization for list buildingSource
referral-programReferral and affiliate program design and implementationSource
sales-enablementSales collateral and pitch deck creation for B2B teamsSource
analytics-trackingAnalytics setup and measurement implementationSource
marketing-psychologyBehavioral science application to marketing copy and UXSource
seo-audit-fullDeep technical and content SEO auditing workflow for comprehensive site/page analysisSource
vibe-prospectingLive B2B company and contact data for natural-language lead-list building, prospect enrichment, executive discovery, and multi-step GTM workflowsSource
reddit-automationFind Reddit threads where people are genuinely asking for what you offer, then draft honest, helpful replies with affiliation disclosureSource

NameDescriptionLink
proficiently-claude-skillsAI-powered job search, resume tailoring, and cover letter writingSource
claude-code-job-tailorResume optimization system that analyzes job postings, ranks requirements, and generates tailored PDFs in under 60 secondsSource
resume-tailoring-skillAI-powered resume tailoring optimized for specific job descriptions while maintaining factual integritySource
claude-resume-kitExtract experience once, generate tailored LaTeX resumes per JD with anti-fabrication controls and multi-perspective critiqueSource
Resume-BuilderAI-powered resume and cover letter generator with dual ATS + HR scoring for any professionSource
ResumeSkillsResume optimization, ATS scoring, interview prep, and strategic job search skills collectionSource

🔐 Security & Testing

NameDescriptionLink
Trail of Bits Security SkillsSecurity skills for static analysis with CodeQL/Semgrep, variant analysis, code auditing, and fix verificationSource
webapp-testingUI test automation using PlaywrightSource
ffuf_claude_skillFuzz test web apps with FFUF + ClaudeSource
defense-in-depthImplement multi-layered testing and security best practicesSource
systematic-debuggingStructured debugging with hypothesis testing and validationSource
testing-anti-patternsIdentify and prevent testing anti-patternsSource
testing-skills-with-subagentsVerify new skills using subagents and test cyclesSource
verification-before-completionRun verification checks before closing tasksSource
condition-based-waitingUse logical conditions to control test flow timingSource
varlock-claude-skillSecure environment variable management ensuring secrets never appear in sessions, terminals, logs, or git commitsSource
security-fuzzingEssential fuzzing payloads for SQL injection, command injection, NoSQL injection, and LDAP injection testingSource
security-payloadsSpecialized attack payloads for XSS vectors, XXE, template injection, and file upload bypass testingSource
llm-testingAI/ML security testing including bias detection, data leakage, alignment testing, and adversarial resistanceSource
accessibility-skillsAccessibility guidance for accessible interfaces, including forms, keyboard use, color contrast, and light/dark modesSource
security-hardeningPortable defensive security workflow covering secure design and review, OWASP, MCP security, prompt injection, incident response, and hardeningSource

⚙️ Utility & Automation

NameDescriptionLink
writing-great-skillsReference for writing and editing skills well — the vocabulary and principles that make a skill predictableSource
git-guardrails-claude-codeInstall guardrails to block dangerous git commands before executionSource
setup-pre-commitConfigure Husky pre-commit hooks with lint-staged, formatting, and quality checksSource
file-organizerClean up file structures, rename documentsSource
invoice-organizerParse and categorize invoicesSource
raffle-winner-pickerPick winners using secure randomnessSource
skill-creatorBuild your own skill interactivelySource
template-skillA starting template for new skillsSource
gardening-skills-wikiMaintain the skills wiki, ensuring naming consistency and metadata qualitySource
pulling-updates-from-skills-repositorySync and pull the latest skill updates from repositoriesSource
cc-devops-skills31 DevOps skills covering IaC, CI/CD, Kubernetes, observability, and scripting automationSource
opentelemetry-skillsInstrumentation, telemetry-quality, collector, semantic-convention, and OTTL guidance for observable applicationsSource
devops-claude-skillsDevOps workflow skills for Terraform, K8s troubleshooting, AWS cost optimization, and GitOpsSource
aws-well-architected-reviewReusable playbooks for applying the AWS Well-Architected Framework to architecture reviews and improvement plansSource
firecrawl-cliOfficial Firecrawl skill for scraping, crawling, searching, and mapping the web via CLISource
hermes-tweetNative Hermes Agent plugin skill for X monitoring, account research, trend checks, and approval-gated social actions. Xquik is an independent third-party service. Not affiliated with X Corp. "Twitter" and "X" are trademarks of X Corp.Source
replicateOfficial Replicate skill for discovering, comparing, and running AI models via APISource
agentsys36 workflow automation skills for profiling, code review, AI consultation, release management, and drift analysisSource
better-i18nOfficial i18n skill for internationalization best practices, translation workflows, and localization automationSource
migrate-to-shoehornMigrate test assertions to @total-typescript/shoehorn patternsSource
scaffold-exercisesScaffold structured exercise directories with sections, problems, solutions, and explainersSource
edit-articleAssist with article editing and refinement workflowsSource
obsidian-vaultWork with and organize notes in an Obsidian vaultSource
obsidian-cliUse Obsidian CLI workflows to automate vault operations and content managementSource
wizardGenerate an interactive bash wizard that walks a human through a manual procedure — opening URLs, capturing values, confirming each stepSource
setup-ts-deep-modulesWire dependency-cruiser into a TypeScript repo so each package is a deep module with a small public surfaceSource
claude-handoffHand the current conversation off to a fresh background agent that picks up the work immediatelySource
find-skillsOfficial Vercel skill that discovers and installs agent skills from the ecosystem when you ask "is there a skill for X"Source

| danshari-skill (断舍离.skill) | Audits every skill you have installed and archives the ones your current model, harness, or MCP servers already cover — three-way triage, blind-test evidence before any removal, and one-command restore instead of rm | Source | | sijiao-skill (私教.skill) | Distils any skill you want to learn into a stateful tutor — it remembers where you are, sets and grades exercises at your level, and schedules spaced review, with the ceiling honestly capped at competent | Source |

Getting Started

Install across supported agents

Skills use the portable SKILL.md format. The open-source Skills CLI installs compatible skills for 70+ coding agents, including Claude Code, Codex, Z.ai's ZCode, Cursor, GitHub Copilot, Gemini CLI, and OpenCode. Copy a source link from a row above, then list and install only the skill you want:

# Inspect the skill names available from a source repository.
npx skills add https://github.com/anthropics/skills --list

# Try a skill without installing it (replace the agent ID as needed).
npx skills use https://github.com/anthropics/skills --skill claude-api --agent claude-code

# Install a specific skill for the current project (the default scope).
npx skills add https://github.com/anthropics/skills/tree/main/skills/claude-api --agent claude-code

# Make it available in every project, or install it for another supported agent.
npx skills add https://github.com/anthropics/skills/tree/main/skills/claude-api --agent codex --global
ToolSkills CLI agent IDProject pathGlobal path
Claude Codeclaude-code.claude/skills/~/.claude/skills/
Codexcodex.agents/skills/~/.codex/skills/
Z.ai ZCodezcode.zcode/skills/~/.zcode/skills/
Cursorcursor.agents/skills/~/.cursor/skills/
GitHub Copilotgithub-copilot.agents/skills/~/.copilot/skills/
Gemini CLIgemini-cli.agents/skills/~/.gemini/skills/

For ZCode, you can also use Settings → Skills → Import to scan and import skills already installed for Claude Code, Codex, and other supported agents. For a tool not listed here, place the skill directory containing SKILL.md in that tool's documented project or user skills folder, then refresh/restart the tool.

Use the interactive prompts to choose the installation method and review the files. Do not use --all or -y unless you have inspected and trust every selected skill. To check installed skills or refresh them later:

npx skills list
npx skills update

Let an agent tailor the selection

Paste this into Claude Code, Codex, ZCode, or another coding agent from inside your project:

Audit this repository read-only and recommend a small, personalized set of skills from https://github.com/abubakarsiddik31/claude-skills-collection. Match the recommendations to the stack, workflows, and current pain points. Exclude NSFW content and programming-language-specific skills. For each recommendation, show the source URL, the exact skill name, prerequisites, data/credential implications, and why it fits. Inspect the skill files before recommending installation. Do not install, change configuration, or use non-interactive flags until I explicitly approve the exact list.

After you approve the list, ask the agent to install the selected skills at project scope for shared team workflows or global scope for a personal toolkit. Name your host explicitly—for example claude-code, codex, or zcode—and pin a source revision when reproducibility matters.


Skill Quality Standards & Template

Writing a great skill starts with a well-structured SKILL.md file. AgentSkills.io provides the canonical specification for the Claude Skills format, including validation rules and a detailed checklist. Anthropic also publishes best practices for skill authoring at platform.claude.com.

Minimal SKILL.md Template

Every skill should include a SKILL.md with proper YAML frontmatter at the top:

---
name: my-skill-name
description: |
  Generates weekly status reports from project data.
  Use when the user asks for a summary of recent activity.
---

The body of SKILL.md (below the frontmatter) contains the instructions Claude follows when the skill is activated.

Key Quality Criteria

  • name -- Use kebab-case, 1-64 characters (e.g., csv-data-summarizer, not CSV Data Summarizer).
  • description -- Write in third person. State what the skill does and when to use it. Keep it under 300 characters.
  • Body length -- Keep the instruction body under 500 lines. Long skills should be split into focused sub-skills.
  • Progressive disclosure -- Start with the most common usage, then cover edge cases. Claude reads top-down, so put the critical path first.
  • No secrets in source -- Never embed API keys, tokens, or credentials in SKILL.md. Use environment variables or MCP configuration instead.

Reference Implementations


Contributing

Have a skill to add?
Open a pull request or submit your repo link in an issue with:

  • Name of the skill
  • Short description
  • Category
  • Link to the source

License

This repo lists and links to skills under various licenses.
Please refer to each linked repository for license terms.


Significant stargazers

Trevin Chow

571 followers · starred Oct 2025

Agustin Rivera

41 followers · starred Dec 2025