Coding standards, frameworks, refactoring, and code review (16.8K skills)
Build full-stack React applications with Next.js, organize multiple apps and shared packages with Turborepo, and add RemixIcon icons. The skill covers rendering and data-fetching patterns, build pipelines, caching, and common project structures for single apps and monorepos.
Build browser based 3D experiences with Three.js, from scenes and animation to shaders, materials, games, data visualizations, and product configurators. The skill covers rendering, asset loading, interaction, and performance techniques for WebGL and WebGPU projects.
Use this skill to add payment flows with SePay, Polar, Stripe, Paddle, or Creem.io. It guides checkout, subscriptions, webhook handling, VietQR payments, licensing, and order management across providers, with platform specific references and helper scripts.
Follow a consistent approach to building React and TypeScript interfaces, from components and feature folders to data fetching, routing, and MUI styling. The guide covers Suspense-based loading, lazy loading, performance practices, and TypeScript conventions.
Work with MongoDB and PostgreSQL across schema design, queries, aggregations, indexes, migrations, backups, permissions, and performance checks. Get practical guidance for common development and production database tasks in both document and relational databases.
Handle code review feedback by checking its technical accuracy before making changes, and request reviews for major work. Follow verification steps so claims about passing tests or completed tasks are backed by fresh evidence.
Use Puppeteer CLI scripts to automate browser tasks and inspect web pages. The skill can navigate, click, fill forms, capture screenshots, collect page data, monitor console and network activity, and measure performance including Core Web Vitals. Results are provided as JSON.
Use this skill to plan and build backend systems, from API design and authentication to database tuning, security, testing, deployment, and monitoring. It offers technology choices and practical checklists for turning requirements into production-ready services.
Stress a design mentally by considering extreme volumes, speeds, user counts, durations, and failure rates. This helps reveal scaling limits, concurrency risks, memory growth, and whether an architecture needs asynchronous work, persistence, or stronger error handling.
Develop Shopify apps, extensions, and storefront themes. The skill covers store data APIs, checkout and admin customizations, POS extensions, webhooks, billing, and Liquid templates, helping shape tools and shopping experiences for Shopify stores.
Apply layered checks to data as it moves through a system. The skill guides you to validate inputs, verify business rules, guard against unsafe environment-specific operations, and log context so invalid data is caught at multiple points.
Use this skill to discover tools, prompts, and resources exposed by configured Model Context Protocol (MCP) servers, select capabilities for a task, and run tools through a command-line interface or Gemini CLI. It also describes multi-server client patterns and configuration.
Investigate bugs and test failures through root cause tracing, hypothesis testing, and layered validation. The skill helps identify where invalid behavior begins, apply safeguards against recurring issues, and verify results with fresh evidence before declaring a fix complete.
Build MCP servers, which give AI agents tools for external services, using Python or TypeScript. The skill guides tool design, input validation, error handling, code review, and evaluations based on realistic tasks.
Create or update skills that give Claude specialized procedures, knowledge, and tool instructions. The guide covers defining use cases, organizing SKILL.md and supporting files, testing scripts, and validating a distributable skill package.
Plan and improve lightweight iOS experiences that users can open without installing the full app. This skill covers App Clip use cases, discovery paths, card content, handoff to the full app, implementation checks, and usage measurement.
Diagnose why an iOS or Android app was rejected, map the review notice to a likely policy issue, and prepare a focused fix plan, reviewer response, and resubmission checklist. It also helps decide when an appeal or expedited review is appropriate.
Use crash reports and stability data to understand an app’s crashes, identify the issues affecting the most users, and prioritize fixes. The skill also covers Crashlytics setup, reading crash reports, symbolication, and tracking crash trends across releases.
Find and install reusable HyperFrames blocks and effect components, then integrate them into an existing composition. Configure block timing and layer order, merge component HTML, CSS, and JavaScript, and adjust installation paths.
Write GSAP animations for HyperFrames compositions, from element entrances to coordinated sequences. Use timelines to organize timing, control playback, adapt motion to screen size and reduced-motion preferences, and choose animation properties suited to efficient rendering.
Use this skill to review LoongCollector self-monitoring code in C++ and Go. It covers metric names, labels, metric registration and updates, and severity-based alarms, helping agents implement or check monitoring changes against project conventions.
Consult this reference for OMC agent roles, tools, workflow skills, team pipeline stages, and commit message trailers. It helps users find the right orchestration commands and understand how work moves through planning, execution, verification, and fixes.
Use this skill to turn a Git diff or supplied change context into a clear Conventional Commits message. It selects a change type and optional scope, explains the reason and impact, and can include issue or breaking change references.
Build LoongCollector’s C++ core and Go plugins, run C++ unit tests, or create end-to-end and release images. The guide helps choose CI-aligned Docker builds or local CMake workflows, with commands and setup notes for each target.