Escape generic AI-generated UIs. Forces a bold, distinctive design direction before any code is written.
By agent
Top Skills for Claude Code
Anthropic's terminal CLI for coding agents. Filesystem skills, plugin marketplace, subagents.
About Claude Code
Claude Code is Anthropic's terminal-first coding agent — a CLI that drives a working Claude session with full filesystem access, shell execution, multi-agent workflows, and native SKILL.md support. Skills installed in Claude Code run with the broadest available capability surface: read and write files, run Bash commands, drive a preview server, query MCP servers, and dispatch subagents through the Agent and Workflow primitives.
The install model is simple — Skills live as folders inside `~/.claude/skills/` or the project's `.claude/skills/`, and the agent loads each Skill's SKILL.md frontmatter at session start to decide when to trigger. Plugins bundle several Skills together for one-shot install via the plugin marketplace. Most Anthropic-authored Skills (code-review, simplify, verify, init, security-review, docx, pdf, pptx, xlsx, mcp-builder, skill-creator) ship pre-installed.
Claude Code is the deepest target for any Skill — the longest-running agent integrations, the most thoroughly tested Skills, and the highest install counts almost always trace back to Claude Code first. We rank Claude Code Skills with provenance, install count, install ergonomics, and recency as the dominant signals.
Verified for this agent
Best Skills for Claude Code
Every skill below is verified to work with Claude Code. Ranked by composite score.
One skill covering every document format the category exists for — including scanned PDFs and RAG chunking — from the maintainers of the parser itself.
Write HTML, render video. Agent-native, deterministic, no build step.
1,234+ skills in one bundle. Single install across AI coding assistants.
Programmatic video done right. Idiomatic Remotion patterns auto-activate when detected.
Structured multi-step development. Brainstorm → spec → plan → build → review → merge — TDD baked in.
Microsoft's first-party Playwright skill, invoked through the CLI — the reference implementation for driving a real browser, with no credentials required.
Firebase-authored. Init, configure, and one-step deploy to Firebase Hosting / App Hosting.
Local browser with no credentials, or a hosted Browserbase session with CAPTCHA solving and residential proxies when the target is protected.
Diff-aware adversarial QA: it reads the git diff and tests only what changed, across correctness, accessibility, layout and UX.
Diagrams generated from live infrastructure rather than from a prompt — it reads the actual resource group and draws what is really there.
Anthropic's own sales stack: research → outreach → call prep → pipeline, nine skills in one plugin.
Stripe's own integration playbook: pick the right API, set up Connect and billing, and handle keys and webhooks securely.
Provision a Postgres, Redis, auth, or LLM provider through the Stripe Projects catalog rather than signing up by hand.
Resend's own API skill: send transactional and batch email, handle inbound webhooks, and manage domains, templates, and keys.
Resend's deliverability playbook: SPF/DKIM/DMARC, spam avoidance, bounce handling, and CAN-SPAM/GDPR/CASL compliance.
Build cross-client HTML emails as React components, with a visual editor and Resend-ready rendering.
Give an agent a webhook-driven inbox with the sender-allowlist and content-filtering patterns that block prompt injection from inbound email.
The marketplace architecture decision, made with a compatibility matrix that blocks invalid Connect configurations instead of describing them.
Multimodal extraction across documents, images, audio and video through one Azure client — Microsoft-authored, with an explicit trigger list.
Batch document translation that preserves layout — the part that breaks when you translate a PDF by extracting text and re-flowing it.
Chrome's own team, giving agents the DevTools protocol: drive it, inspect it, debug it.
Stripe's answer to agentic checkout: one-time payment credentials so an agent can buy without ever seeing the underlying card.
The only vendor-official skill from a major diagramming product — and it emits editable .drawio files rather than a throwaway image.
Vendor-official voiceover generation across 70+ languages, from the company that defined the category. Ships with evals, which almost no skill does.
Transcription and subtitle generation as a first-class skill — the media job agents are asked for most and could not previously do well.
Localise video while keeping the original speaker's voice — and correct the transcript before regenerating, rather than accepting the first pass.
Stripe's version-upgrade guide: move API versions and SDKs without breaking your integration.
Find vendors and partners for a job-to-be-done — and, where supported, let the agent purchase and consume the service.
A self-updating vector-database skill: it fetches current Qdrant guidance at run time instead of shipping a snapshot that rots.
One skill across Postgres, MySQL, SQL Server, MariaDB and SQLite — and its actual job is forcing explore-before-query on real schemas.
Local document extraction with no API key — read the file, not the whole file.
Hookdeck's webhook playbook: 40 provider-specific skills covering signature verification, idempotency, and event handling — each with runnable examples.
Composio's official skill for connecting agents to 1000+ external apps: auth configs, per-user connections, tool execution, and triggers.
Generative video from the vendor developers name first, as a skill set rather than an SDK crawl.
1,000+ SaaS integrations as one skill. OAuth lifecycle, scoped credentials, structured actions.
Supabase done right. RLS policies, edge functions, realtime, auth — in idiomatic patterns.
Neon serverless Postgres done right — branching, pooling, and scale-to-zero, from Neon.
The generalist Postgres skill the ecosystem was missing — schema, indexing, extensions and migrations, from a Postgres company.
Zero-downtime Postgres migrations, with the locking behaviour spelled out — which ALTER TABLE takes an AccessExclusiveLock and what to do instead.
Schema design decisions made explicitly: data types, key strategy, JSONB versus columns, and when partitioning earns its complexity.
First-party HubSpot CRM operations for agents: lookups, deals, quote-to-cash, reporting.
Open-source marketing stack for AI coding agents. SEO, content, email, ads, analytics in one bundle.
The deepest MoR skill coverage after Stripe: 17 skills from checkout to go-live.
SaaS email as skills: Loops API, CLI, LMX markup, deliverability practice.
Box as an agent surface, plus the contract-review workflows nothing else in the catalog covers.
Full-stack WorkOS implementation with auth provider migration paths.
Drop-in WorkOS widgets with backend token wiring across every major framework.
Web scraping, search, and browser automation built to be agent-friendly.
PlanetScale's schema branching and index-aware queries, packaged for agents.
shadcn/ui context: components, patterns, and the right composition by default.
Prisma migrations and queries by the book. Index-aware, migration-safe, type-correct.
Caching without the bugs. Invalidation, write-through, cache-aside, distributed locks.
Paddle's own integration playbook: ten skills for Checkout, webhooks, pricing pages, and subscription lifecycle on the merchant-of-record platform.
Twilio's official Developer Kit: 55 skills across SMS, WhatsApp, Voice, Verify, SendGrid, and compliance — plus the Twilio MCP. Public beta.
The SQLite slot, filled by Turso — and unusually, a skill that actively steers the agent off the stale API surface web search returns.
Slack's own Block Kit skill — and the rare vendor skill that needs no token, because it validates against Slack's public API.
Transactional email the Postmark way: send, inbound, templates, webhooks.
Self-contained HTML artifacts with React + shadcn/ui — no build step, just paste and run.
Extract, fill, merge. PDF manipulation with bundled scripts so Claude does not waste tokens generating them.
Real exploits, no false positives. 96.15% exploit success across 50+ vuln types.
Sentry the right way. Source maps, releases, breadcrumbs, performance, structured tags.
Semgrep's own scanning skill: run the industry-standard SAST engine across 30+ languages, and author custom rules for your codebase's patterns.
A vendor-maintained rule corpus for secure code review — one reference file per vulnerability class, from SQL injection to Terraform misconfiguration.
The vector-database skill that ships executable scripts rather than prose — the agent runs the query instead of describing it.
Pre-handoff a11y audit that catches what design QA misses.
Deterministic web control. ref-based clicks/fills, no flaky selectors, parallel sessions.
MongoDB aggregations and indexing without the foot-guns.
Stripe done right. Checkout, subscriptions, webhooks, idempotency, metered billing.
Notion API: page/database CRUD, block hierarchies, rich text, schema sync.
Shopify Storefront API: cart, checkout, metafields, customer accounts, Hydrogen patterns.
GitGuardian's own secret-detection skill — scans history, images and packages, and ships a written remediation doctrine rather than just a list of hits.
All of Railway in one skill: provision, deploy, debug — listed in the official plugin marketplace.
The geospatial gap, filled by the vendor that defines it: data on a map, done properly.
ADR authoring + design review. Trade-offs, consequences, alternatives.
Disciplined debugging loop. Reproduce, isolate, diagnose, fix.
The pre-ship checklist that catches the embarrassing stuff.
Technical writing that engineers will actually read and update.
Incident triage + comms + blameless postmortem in one harness.
System design from requirements. APIs, data, boundaries, scale.
Honest tech-debt audit with a prioritized cleanup backlog.
Layered test plan sized to the change's risk profile.
Single seed prompt → cohesive token system. Tokens that look intentional, not generated.
Test local webapps with Playwright. Generation, run, report — end-to-end.
Review for reuse, quality, efficiency. Catches over-abstraction and dead code before merge.
The multi-channel notification slot nothing else covers: push, in-app and chat alongside email, with copy guidance per channel. No credentials required.
The widest format matrix of any conversion skill we catalogue, and it executes the conversion rather than describing it.
FHIR to prior-auth to ICD-10 coding: the official payer/provider/pharma toolkit.
One-command productivity bootstrap. Memory, tasks, dashboard.
Cited research reports from fanned-out web search with adversarial verification.
Real feature verification — run the app, drive it, prove it works.
Bootstrap the right plugin stack for your role in one guided pass.
100+ accessibility and UX rules from Vercel, applied as a structural audit on your UI code.
57 React/Next.js performance rules, applied as a structural review on your code.
Security audits run by a real security firm. CodeQL + Semgrep + audit workflows.
Strip the AI tells. Multi-pass editing for content that reads opinionated, not generated.
Next.js App Router by the book. Server components, streaming, parallel routes.
Ship to a live Vercel URL from chat — auto-detects the framework, returns a claim link.
OWASP-published review, SCA, secrets and threat-modelling skills — the most citable provenance available for security review vocabulary.
The remediation half of security work: scan, fix, validate the fix, open the PR. Batch mode handles a whole backlog of CVEs in one pass.
Senior-designer feedback on demand. Hierarchy, usability, consistency, motion.
Turn a final design into a spec engineers can implement directly.
Design-system hygiene at speed. Audit, document, extend.
Microcopy that respects the user. Errors, empty states, CTAs, onboarding.
Lock generated UIs to a brand. Anthropic's own guidelines, formalized as a skill.
Word documents end-to-end. Create, edit, comment, track-change.
Excel spreadsheets with real formulas, charts, and formatting — not flat CSV exports.
PowerPoint decks with layouts, charts, speaker notes — programmatic from prose.
Build new MCP servers. The official scaffold + design walkthrough.
Build new skills interactively. Output is a clean SKILL.md ready to publish.
Always-current Claude API reference and SDK docs. Bundled with Claude Code.
p5.js algorithmic art with particle systems and procedural geometry.
Visual art from prose. PNG and PDF outputs that look intentionally designed.
Animated GIFs sized and optimized for Slack reactions.
End-to-end blog pipeline. 30 sub-skills, 5 agents, a 5-gate quality contract that blocks publish below 90/100.
Web search + 36 authoritative data sources. SEC, PubMed, ChEMBL, FRED — agent-grade research.
Internal comms that read consistent. Newsletters, status reports, FAQs, updates.
Multi-author edits with suggestion mode and conflict resolution.
Actually renders Mermaid to SVG or ASCII instead of emitting text and hoping something downstream draws it. Bundles the render scripts.
The official entry point for biomedical research workflows in Claude Code.
The largest payment processor, first-party: checkout to disputes to payouts.
The anti-bot problem, handled: proxies, SERP, and structured extraction at scale.
Pre-merge security sweep on the current branch's diff.
The canonical Anthropic code review. Effort levels from quick to ultra.
Bootstrap a CLAUDE.md so future sessions start informed.
Edit settings.json safely. Hooks, perms, env, behaviors.
Rebind keys and add chords without reading the keymap docs.
Cuts the permission-prompt friction by learning your safe-tool patterns.
Cron for remote Claude Code agents. Set it and forget it.
Boolean prop hell → compound components. Cleaner React component APIs by construction.
Headless browser as an agent tool. Navigate JS-heavy pages, fill forms, extract content.
20 SEO + GEO skills in one bundle. Keyword research, audits, rank tracking, AI search optimization.
Scientific computing + dataset APIs. Reproducible research workflows for agents.
232+ skills across engineering, marketing, product, compliance, advisory.
Expo workflows: navigation, EAS, OTA, native modules.
In-process analytical SQL over CSV, Parquet, and Excel — no warehouse, no server.
Mermaid diagrams from prose with self-validating syntax. Sequence, flowchart, ER, state.
LinkedIn posts that read human. Hook patterns, voice modeling, carousel scripts.
The full outbound motion in skills: qualify, map, outreach, handle objections, propose.
A full SEO platform as skills: briefs, audits, backlink gaps, AI-search share of voice.
The heavyweight audit machine: 88 evidence scripts so SEO findings come from data, not vibes.
Data-driven document generation: one template, a thousand filled documents.
Run the marketplace's scrapers, or turn your own scraper into a deployable Actor.
Turn raw research into themes, segments, and a prioritized next-steps list.
Research-ops in a skill. Plan the study, run it, synthesize the findings.
Brand-voice QA before content ships. Severity-ranked findings with rewrites.
Launch-ready campaign brief in one pass. Calendar, channels, metrics included.
Battlecards on demand. Positioning gaps, content holes, threat assessment.
Channel-correct content with SEO and CTA awareness baked in.
Channel-specific draft + headline options. SEO-aware out of the box.
End-to-end drip campaigns. Copy + timing + branching + benchmarks.
Executive-ready performance reports with next-period priorities.
Full SEO health check with quick wins vs strategic investments split.
Workplace memory that actually understands your team's shorthand.
Lightweight task tracking in a single markdown file.
TDD with red/green/refactor as a concrete TypeScript walkthrough.
Netlify-authored deploy flow. Auth, site linking, Deploy Previews, production promotes.
Cloudflare-authored. Workers and Pages deploys, Wrangler config, KV/D1/R2 from chat.
The standalone FHIR skill: modeling, validation, SMART on FHIR — without the full healthcare plugin.
Brand-checked design output at volume — the production half of design work.
32+ marketing skills in one bundle. CRO, copywriting, SEO, analytics, email, growth.
Structured PR reviews with severity-tagged findings — bugs, security, perf, style.
Repo-to-live-URL on GitHub Pages: Actions workflow, custom domain, Jekyll, troubleshoot.
Deploy, debug, and scale on Render: Blueprint generation with validation built in.
A named design canon for charts, not another plotting wrapper.
Tailor a plugin to your org's stack in one pass.
Scaffold a new plugin from scratch with guided prompts.
Excalidraw JSON diagrams from prose, with rendering self-checks.
A HubSpot admin in a plugin: audits, dedup, scoring, lifecycle hygiene — 37 skills deep.
Terraform expertise on tap. State inspection, module validators, safety-first IaC.
Senior DevOps engineer persona — CI/CD, containers, blue-green/canary, IaC across clouds.
HTML slide decks from prose. Easy to embed, version, and edit later.
C4, deployment, sequence-of-events architecture diagrams generated from natural-language descriptions.
Copywriting + SEO bundle. SEO audits, schema markup, programmatic-SEO patterns, Ogilvy-school copy.
Astro from scaffold to live site — content collections, search, i18n, multi-host deploy.
Drive the iOS Simulator from the agent loop. Build and test iOS apps.
Favicons, app icons, social cards — generated at all the right sizes.
Multi-agent orchestration. Spawn parallel subagents and merge their outputs reliably.
Kubernetes hardening by the book. NetworkPolicies, RBAC, OPA, service mesh.
Playwright patterns: clicks, fills, waits, retries. The do-it-right defaults.
D3.js patterns for interactive viz that does not look stock.
Full-stack SEO: 19 sub-skills, 12 subagents, schema, GEO/AEO, backlinks, local.
ffuf web fuzzing for authorized pentests. Common modes, payloads, and gotchas.
Operate a Lemon Squeezy store from the agent: customer lookup, refunds, discount codes, receipts, and a sales dashboard — six support-and-revops skills.
FAQ
Frequently asked
How do I install a Skill on Claude Code?
Drop the Skill folder into `~/.claude/skills/` (user-wide) or `.claude/skills/` (project-scoped), or run the plugin install command. The agent picks up new Skills on next session start.
Which Skills ship with Claude Code by default?
code-review, simplify, verify, init, security-review, run, the Anthropic document Skills (docx, pdf, pptx, xlsx), mcp-builder, skill-creator, and several others. The full list is in your session's available-skills block.
Can Skills run Bash commands?
Yes — Claude Code runs shell commands with your approval. Skills that automate deploys, run tests, or invoke external tools all rely on this.
Do Skills work with subagents?
Yes — Skills can be invoked from within the Agent and Workflow primitives, including in parallel across multiple subagents.
Is Claude Code free?
Claude Code itself is free; usage is billed against your Claude API or Claude.ai subscription. Some plugins and Skills are also free; others are paid — check the Skill page.
Switch agent