Claude Skills
The open format is called Agent Skills and works in Claude Code, Codex, Cursor and other agents — most people know it as Claude Skills.
Every Agent Skill we could find on GitHub, deduplicated by content. 79 551 files from 1 750 authors, of which 61 898 are unique — the rest is the same skill repackaged into someone else's repository. For each one: what it weighs in tokens, whether it ships runnable scripts, and which MCP servers it needs.
Designing LLM-optimized folder structures. Audits and restructures directories for context efficiency, progressive disclosure, and prompt cache performance. Don't use for general repo structure (Grove), config audit (Hone), or skill generation (Sigil).
Implementing pure-native app features for iOS (Swift 6.3 + SwiftUI + Liquid Glass), Android (Kotlin 2.4+ + Jetpack Compose + Material 3 Expressive), and macOS desktop (SwiftUI for Mac + AppKit interop, scenes/menu bar/document apps, sandbox + notarization + Sparkle, Mac HIG). Builds production features with @Observable/Swift Concurrency, Compose Strong Skipping, SwiftData/Room, Passkeys, Privacy Manifest, App Intents, Foundation Models/Gemini Nano, and store-compliance staged rollout. Also runs the agent visual loop for screen implementation and visual debugging against a reference design (Xcode MCP RenderPreview, XcodeBuildMCP, mobile-mcp, screenshot diffing). Use when building production iOS/Android/macOS features or iterating a native screen until it matches a design. Not for cross-platform (RN/Flutter/KMP/CMP), porting design (Port), prototypes (Forge), mockup-to-code (Pixel), automating existing Mac apps via AppleScript/JXA (Hearth), test suites (Snap/Voyager), or web (Artisan).
Enumerating failure modes via pre-mortem analysis. Systematically identifies failure scenarios for plans, designs, and features, scoring them with RPN/AP. Does not write code.
Designing and evaluating AI/ML systems across prompt engineering, RAG design, LLM application patterns, AI safety, evaluation frameworks, MLOps, and cost optimization. Use when designing AI/ML pipelines, RAG architectures, prompt strategies, evaluation harnesses, or LLM cost models.
Improving usability, interaction quality, cognitive load reduction, feedback design, and a11y compliance. Use when improving UX usability or interaction feel.
Navigating project delivery status as a PdM-style read-only navigator. Reconciles planned scope (specs/issues/roadmap/PRD) against implemented code to produce feature inventories, unimplemented-feature lists, roadmap rollups, and WBS views. Use for "what's built / what's left / where are we". Don't use for code comprehension (Lens), priority scoring (Rank), spec authoring (Scribe), AC conformance (Attest), or live task execution (Sherpa).
Running autonomous loops for nexus-autoloop. Generates script sets from goals, designs operation contracts, audits live loops, and recovers state — delivering end-to-end runners that complete reliably.
Designing GitHub Actions workflows in depth — covering trigger strategy, security hardening, performance optimization, PR automation, and Reusable Workflow design. Use when new GHA workflow design or advanced optimization is needed.
Generating pixel-accurate HTML/CSS code from image mockups (PNG/JPG/screenshots) and performing visual verification for faithful reproduction. Use when mockup-to-code generation is needed.
Role-playing as end users to generate authentic feature requests, surface unmet needs, and challenge team assumptions as a synthetic user advocate. Don't use for real feedback analysis (Voice) or UI evaluation (Echo).
Designing web-to-iOS/Android porting strategies. Produces feature parity matrices, native architecture maps, platform-UX adaptation, data/auth/CRDT strategies, BFF redesigns, and Strangler-Fig phased roadmaps from React/Vue/Svelte/Angular SPAs (RSC/SSR included). Optionally proposes a hybrid path (pure-native UI + KMP shared logic). 2026 spec aware (Swift 6.3 / Compose 1.11 / Privacy Manifest / 16KB / Passkey / DMA / EAA). Use when designing web-to-native ports. Not for cross-platform UI (Native — RN/Flutter), same-language framework migration / dependency upgrades / modernization (Shift), legacy archaeology (Trail `static-rules`), or pure-native impl (Native).
Implementing internationalization (i18n) and localization (l10n). Extracts hardcoded strings to t() functions, integrates Intl API for date/currency/number formatting, manages translation keys, and adds RTL layout support.
Cleaning up the skill ecosystem by auditing the agent roster for overlap, redundancy, and inactivity; proposes merge candidates and sunset plans with evidence + archive instructions. Propose-only (no execute). Use when the ecosystem needs portfolio cleanup, agent consolidation, or sunset planning. Not for individual skill quality (Architect IMPROVE), strategic ecosystem direction (Darwin), code-level YAGNI (Void), file-level dead code (Sweep), or SKILL.md format audit (Gauge).
Writing user-facing UX text including microcopy, error messages, voice and tone design, onboarding copy, and accessibility text. Use when UX writing or content strategy is needed.
Defining KPIs, designing tracking events, and specifying dashboards. Covers North Star Metric, funnel analysis, cohort analysis, and test-intelligence dashboards (flake rate, regression timeline, mutation-overlaid coverage — absorbed from vista). GA4/Amplitude/Mixpanel/PostHog integration. Use when metrics or test-telemetry dashboards are needed.
Adding JSDoc/TSDoc, updating READMEs, replacing any types with proper definitions, and adding high-value comments to complex logic. Use when documentation gaps, unclear intent, or type safety improvements are needed.
Integrating OWASP ZAP/Burp Suite/Nuclei, planning penetration tests, executing DAST, and scanning for vulnerabilities. For dynamic security testing, pentesting, or runtime vulnerability validation. Complements Sentinel static analysis.
Adding edge-case tests, repairing flaky tests, and improving coverage. Use when test gaps need filling, reliability needs raising, or regression tests need adding. Multi-language support (JS/TS, Python, Go, Rust, Java).
Brainstorming interactively to deepen ideas via iterative dialogue using four modes (Expand/Propose/Evaluate/Subtract). Does not write code. Don't use for decisions (Magi), feature specs (Spark), or single-shot reframing (Flux).
Quantifying priority by scoring and ordering competing items using ICE/RICE/WSJF/MoSCoW/Cost of Delay/Kano frameworks. No code. Use when prioritizing features/bugs/initiatives, ranking by ICE/RICE/WSJF/etc., or arbitrating Must-have vs Should-have at MVP scoping.
Orchestrating multi-session parallel execution using Claude Code Agent Teams API and Codex CLI Subagents to launch, manage, and coordinate concurrent task execution across multiple instances. Use when parallel work is needed.
Integrating messaging platforms, developing bots, and designing/implementing real-time communication. Covers channel adapter patterns, webhook handlers, WebSocket servers, event-driven architecture, bot command frameworks. Use when integrating Slack/Discord/Teams bots, designing webhook receivers, or wiring event-driven messaging.
Advising founders YC-style by surfacing the #1 bottleneck via socratic probing, giving pattern-matched honest feedback, detecting founder anti-patterns, and extracting 1-2 week action items. Use when seeking founder-mode advisory. Not for brainstorming (Riff), triadic deliberation (Magi), long-term scenarios (Helm), feature gen (Spark), or impl (Builder).
Analyzing pre-change impact by evaluating risk across vertical (dependency chains, affected files) and horizontal (pattern consistency, naming) dimensions. Does not write code. Use when estimating blast radius before a refactor, audit, or migration — or when a PR's risk surface is unclear.
Designing narratives that tell product and feature use cases as customer-centric stories. Use when customer experience storytelling, scenario stories, or product narratives are needed.
Provisioning infrastructure via cloud IaC (Terraform/OpenTofu/CloudFormation/Pulumi) and local development environments (Docker Compose/dev setup/env vars). Use when IaC design, environment setup, or multi-cloud provisioning is needed.
Investigating bugs via root cause analysis (RCA), reproduction steps, and impact assessment. Investigation-only — identifies why bugs occur and where to fix them, no code. Use when a bug needs RCA, reproduction must be established before fix, or impact radius needs assessment.
Designing database schemas, planning migrations, and authoring ER diagrams. Handles normalization, index strategies, and relation definitions. Use when DB schema design is needed.
Designing search engines and vector DBs for full-text, vector, and hybrid retrieval. Use when full-text search, vector search, or hybrid search design, index optimization, or RAG retrieval layer implementation is needed.
Authoring specifications, design documents, implementation checklists, and test specifications. Handles PRD/SRS/HLD/LLD technical documents, review checklists, and test case definitions. Does not write code. Use when technical documentation is needed.
Guiding workflows by decomposing complex tasks (Epics) into Atomic Steps under 15 minutes each. Manages progress tracking, drift prevention, risk assessment, and timely commit proposals. Use when complex task decomposition is needed.
Designing multi-tenant architectures with tenant isolation strategies, RLS, routing, and scale design for SaaS. Use when designing multi-tenant SaaS systems or tenant isolation.
Analyzing code statically for security flaws: hardcoded secret detection, SQL injection prevention, input validation, security headers, and dependency CVE scanning. Don't use for runtime exploit verification (Probe), general code review (Judge), CI/CD management (Gear), or detection rule authoring (Vigil).
Orchestrating migrations, upgrades, and modernization across frameworks, libraries, APIs, databases, native runtimes, and dependencies. Detects deprecated libraries, suggests native API replacements, runs technology radar, generates codemods, applies incremental strategies (Strangler Fig/Branch by Abstraction), verifies behavioral equivalence, and produces rollback plans.
Automating iOS UI via XCUITest, accessibility-identifier-driven queries, programmatic screenshot capture, and fastlane snapshot pipelines for App Store assets. Authors XCUITest targets with XCUIApplication/XCUIElement/XCUIElementQuery patterns, Page Objects in Swift, status-bar-clean sessions, and xcodebuild+xcresulttool CI. Use when authoring iOS UI tests or store-screenshot pipelines. Not for cross-platform E2E (Voyager — Appium/Detox/Maestro), iOS feature impl (Native), demo recording (Director), or unit tests (Radar).
Verifying system resilience via load testing, contract testing, chaos engineering, and mutation testing. Use when system limit verification, non-functional testing, or reliability validation is needed.
Generating, updating, auditing, and sync-repairing project-specific Claude Code skills. Analyzes the repo stack and conventions, synthesizes Micro or Full skills matched to project patterns, and installs to both .claude/skills/ and .agents/skills/. Use when authoring project-local skills.
Generating AI image-generation code using the Gemini API. Handles text-to-image generation, image editing, and prompt optimization. Use when image generation code is needed.
Proposing new features leveraging existing data/logic as Markdown specifications. Use when brainstorming new features, product planning, or feature proposals are needed. Does not write code.
Generating slides via Marp, reveal.js, or Slidev, designing narrative arcs, and optimizing conference talks with WPM-calibrated timing. Use when creating or pacing presentations.
Designing ETL/ELT pipelines, visualizing data flows, selecting batch/streaming approaches, and architecting Kafka/Airflow/dbt systems. Use when building data pipelines or managing data quality.
Channeling the documented thinking of named notable figures (an 'itako' medium) — applies their mental models, heuristics, and decision-style to your problem as advisory lenses. Don't use for decisions/verdicts (Magi), synthetic user personas (Cast), or fixed founder-mentor coaching (Sage).
Delivering working code fastest via build-first product delivery. Routes through minimum agent chains scoped to the task. S/M scopes skip planning and build immediately. Use when shipping the shortest path to working code, fast-tracking S/M scopes, or compressing time-to-first-PR on small-to-medium features.
Designing scheduling and time-aware logic for cron, timezone/DST, retry/backoff, and business-calendar systems. Use when schedule design is needed.
Detecting unnecessary files, unused code, and orphaned files, and proposing safe deletion. For repo cleanup, dead code removal, or tidying. Don't use for removal execution (Builder), deletion review (Judge), repo structure (Grove), or scope cutting (Void).
Converting repository changes into detailed learning documents. Use when turning diffs into teaching materials, recording design decisions, or creating onboarding materials for new members.
Architecting crawl and scraping systems — distributed crawler topology, URL frontier, politeness, and compliance. Architecture-only (no execution code). Don't use for single-page scraping (Vector) or ETL pipelines (Stream).
Investigating git history, analyzing regression root causes, and performing code archaeology. Time-travels through commit history to uncover truth. Use when git history investigation or regression analysis is needed.
Analyzing session replays, extracting persona-based behavioral patterns, and storytelling UX issues. A behavioral archaeologist that reads the 'why' from actual user operation logs. Collaborates with Field/Echo for persona validation.
Responding to incidents, identifying impact scope, formulating recovery procedures, and creating postmortems. Use when incident response or disaster recovery is needed. Does not write code (delegates fixes to Builder).
Engineering detection rules (Sigma/YARA), mapping detection coverage, designing threat hunting hypotheses, executing Purple Team Blue side, and integrating Detection-as-Code CI/CD. Use when defensive security verification is needed.
Tuning database queries via EXPLAIN ANALYZE, query plan optimization, index recommendations, and slow query detection/fixing. Complements Schema's schema design. Don't use for schema/migrations (Schema), app rewrites (Builder), non-DB performance (Bolt), or unknown root cause (Scout).
Directing UI/UX creative work — complete redesigns, new designs, and trend application. Use when design direction decisions, Design System construction, or orchestration of Muse/Palette/Flow/Forge is needed. Does not write code. Offers an interactive co-design pair mode (decide direction together, one decision at a time).
Automating browsers via Playwright and Chrome DevTools for data collection, form interaction, screenshot capture, and network monitoring. Task completion focus (vs Voyager for E2E testing).
Verifying YAGNI, cutting scope, pruning, and proposing complexity reductions. A 'subtraction' agent that questions the justification for every code, feature, process, document, design, spec, dependency, and config. Does not write code.
Collecting user feedback via NPS surveys, review analysis, sentiment analysis, feedback classification, and insight extraction reports. Use when establishing feedback loops.
Authoring Storybook stories, managing component catalogs, and integrating Visual Regression. Covers UI component docs, visual testing, CSF 3.0/Factories. Supports Storybook 10 (ESM-only) and React Cosmos. Use when authoring Storybook stories, building a component catalog, or wiring Visual Regression into CI.
Authoring E2E tests for web (Playwright/Cypress/WebdriverIO) and native mobile (Appium/Detox/Maestro/XCUITest/Espresso). Covers Page Object design, auth flows, parallel execution, visual regression, a11y, CI, and remote device-farm (BrowserStack/Sauce Labs/AWS/Firebase). Use when authoring E2E suites. Not for unit (Radar), load/chaos (Siege), ad-hoc browser (Vector), or native impl (Native).
Designing workflows and state machines. Use when state transition design, invalid transition detection, Saga patterns, or approval flow design is needed.
Authoring tech blog/article series for note/Zenn/Qiita/dev.to. Not for specs (Scribe) or microcopy (Prose).
Claude Skills — questions
Answers built from the skills we actually parsed.