Claude Skills by ArieGoldkin

Patterns for CLI tool wrapper skills — auto-context, setup verification, output budgeting, error enrichment. Triggers on tool wrapper, CLI wrapper skill, wrapper pattern, auto-context, setup verification, output budgeting

Run tests, linting, and type checking with structured evidence collection. Auto-detects project stack (Node, Python, Rust). Reports quality level: all clear, warnings, failures, or blocked. Optional streak gate requires N consecutive green runs before declaring success. Use when: validating code before commit, checking CI readiness, running all project checks, or collecting evidence for a PR. Triggers on: verify, run checks, quality check, run tests, does it pass, lint, typecheck, pre-commit ...

Force a higher-abstraction explanation of the current code region — modules, callers, glossary vocab. Use when you're unfamiliar with a section of code, need to understand how it fits into the bigger picture, or say "zoom out", "give me the bigger picture", "I don't know this area", "explain at a higher level". Triggers on zoom-out, bigger picture, higher level, broader context, what does this fit into, map of modules.

Headless browser automation via the agent-browser CLI — snapshot + @ref workflow, ~200-400 tokens/interaction vs Playwright MCP. Use when automating a browser, scraping a JS-rendered page, filling forms, testing a web app, logging into a site, capturing screenshots, or introspecting a React app. Triggers on agent-browser, browser automation, scrape, web scraping, fill form, click button, screenshot, E2E test, snapshot refs, react devtools, web vitals

Visual feedback tool for AI agents \u2014 CSS selector capture, DOM context, QA bug reports, and UI review

Decision framework for AI UI \u2014 json-render vs Stitch AI vs Claude Design vs manual components. Triggers on AI UI, generative UI, claude design

Capture JS-rendered SPA content via agent-browser CLI. For when WebFetch returns empty or pages need JS rendering

Design tokens (W3C DTCG) \u2014 CSS custom properties, Tailwind v3/v4, OKLCH color, dark mode, Style Dictionary pipelines

Generate Remotion-based explainer videos for software architecture and process-flow demonstrations. Use when documenting multi-phase pipelines, system architectures, agent dispatch flows, sequence diagrams, or state machines as motion+narration. Triggers on: explainer video, architecture demo, flow video, remotion, animated diagram, render video, system walkthrough, process video, state machine video, sequence diagram video.

Figma MCP workflow \u2014 extract designs, convert to code, Code Connect mappings, design tokens. Triggers on Figma URLs

Distinctive production frontend interfaces avoiding generic AI aesthetics. Anti-generic guidelines and creative direction

UI interaction patterns \u2014 loading states, pagination, modals, drag-and-drop, ARIA live regions, and error recovery

Generative UI with json-render \u2014 Zod-typed catalogs, flat-tree specs, SSE streaming. Triggers on json-render, generative UI

Build interactive HTML playgrounds as self-contained single-file explorers. Visual controls, live preview, and prompt copy. Templates for design, data, concept maps, document critique, diff review, code maps, strategy briefs (roadmaps, option matrices, mermaid diagrams), and decision boards (drag-to-prioritize). Use when: creating an interactive explorer, building a visual tool, prototyping with live controls, or presenting a roadmap/strategy doc. Triggers on: playground, explorer, interactiv...

Convert design prototypes (HTML, CSS, Figma) to production React + TypeScript. Extracts tokens and reusable patterns

Use when the user wants something shown visually right now in the chat — status, progress, comparisons, trade-offs, architecture, blast radius, or an ad-hoc 'show me X' — and wants the picture immediately rather than a file. For a persisted HTML explorer use playground; for diagrams written into a doc use ascii-visualizer. Triggers on quickviz, visualize this, show me visually, draw me a diagram, diagram this, render this as a chart, what does this look like, sketch this out

Responsive design \u2014 container queries, fluid typography, CSS subgrid, View Transitions, scroll animations, Tailwind v4

Manage shadcn/ui components \u2014 add, search, style, compose, debug. Covers registries and presets. Triggers on shadcn

Generate UI designs with Google Stitch AI (Gemini). Create mockups, explore design variants, build multi-page sites, extract design tokens, and chain to production code via DESIGN.md workflow. Use when: creating UI mockups, exploring visual options, generating a design system, or converting designs to React/Tailwind. Triggers on: stitch, vibe design, generate UI, mockup, design variants, design system, UI generation, Stitch AI, visual design, create a design

Production React UI \u2014 shadcn/ui, Radix, CVA variants, React Hook Form + Zod, accessibility, Motion, Tailwind v4

REST, GraphQL, and gRPC API design — endpoint patterns, versioning (URL/header/content negotiation), error handling, pagination, authentication (JWT, OAuth2, API keys), and OpenAPI/AsyncAPI templates.

Use when writing an ASCII diagram INTO A FILE — architecture, workflows, comparison tables, or file trees that must render identically in any font, diff viewer, or CI log. For a quick inline picture in the chat instead, use quickviz. Triggers on ascii diagram, box drawing, architecture diagram, file tree, workflow diagram, monospace diagram, text diagram, diagram in docs

AWS CLI for Lambda deploy/invoke/logs, CloudWatch Log Insights, Secrets Manager, S3 operations, RDS monitoring, and IAM policies. Includes SSO auth, profile management, and platform patterns.

FinOps practices and AWS cost management with Terraform automation. Covers Savings Plans, Reserved Instances, Spot instances, rightsizing with Compute Optimizer, AWS Budgets, cost allocation tags, and storage optimization. Provides actionable templates for budget alerts, tag enforcement, and cost analysis workflows.

AWS networking architecture with VPC, Transit Gateway, and PrivateLink using Terraform. Covers multi-AZ VPC design, CIDR planning, Transit Gateway hub-and-spoke topology, VPC peering, PrivateLink for service exposure, VPC endpoints (Gateway and Interface), security groups, NACLs, and hybrid connectivity patterns.

Structured code review with conventional comments and domain checklists (security, tenant isolation, sensitive data)

Code quality standards — function size, cyclomatic complexity, test coverage, naming conventions, biome/ruff enforcement

Database schema design for SQL (PostgreSQL, MySQL) and NoSQL (MongoDB) — normalization (1NF/2NF/3NF), indexing, constraints, zero-downtime migrations, query optimization, and N+1 prevention.

Databricks workspace and Unity Catalog setup on AWS with Terraform. Use when provisioning Databricks workspaces, configuring Unity Catalog metastores, setting up storage credentials and external locations, managing access control with grants, defining cluster policies, or creating Databricks job workflows. Covers service principal authentication, multi-workspace governance, and Terraform module patterns for Databricks on AWS.

CI/CD pipelines, containerization, Kubernetes, and infrastructure as code patterns for platform deployments. Use when setting up GitLab CI pipelines, configuring Docker multi-stage builds, deploying to Kubernetes or ECS, implementing GitOps with ArgoCD, or managing Terraform/CloudFormation infrastructure. Covers deployment strategies (blue-green, canary, rolling), container security scanning with Trivy, secrets management, and release automation. Handles AWS Lambda deployments, EKS cluster co...

Structured logging, metrics, distributed tracing, and alerting strategies for production systems. Use when implementing CloudWatch/Datadog logging, setting up Prometheus metrics, configuring OpenTelemetry tracing, creating PagerDuty/OpsGenie alerts, or debugging production issues. Covers AWS Lambda Powertools logging patterns, correlation IDs for request tracing, metric dashboards with Grafana, error tracking with Sentry, and SLO/SLA monitoring. Handles log aggregation, anomaly detection, and...

Full-stack performance analysis, optimization patterns, and monitoring strategies for platform. Use when optimizing slow database queries, reducing frontend bundle size, improving Core Web Vitals (LCP, INP, CLS), fixing N+1 query problems, or preparing for high traffic. Covers PostgreSQL query optimization with EXPLAIN ANALYZE, React performance profiling, code splitting and lazy loading, caching strategies (Redis, CDN), and load testing with k6. Handles API latency reduction, memory leak det...

PostgreSQL database operations, schema inspection, Alembic migrations, query optimization, SQLAlchemy 2.0 ORM patterns, and AWS RDS management. For querying, migrations, EXPLAIN analysis, N+1 prevention, and production database work.

Security audit — OWASP Top 10 mitigations, authentication patterns, input validation, compliance, automated scanning

Container-native AWS Lambda deployment patterns with Terraform. Use when deploying Lambda functions from ECR images, setting up ECR repositories with lifecycle policies, configuring VPC-connected Lambdas, or integrating with API Gateway. Covers Docker multi-stage builds optimized for Lambda, X-Ray tracing setup, environment variables and secrets integration, and IAM role configuration. Includes templates for reusable Lambda modules, ECR lifecycle management, and production-ready Dockerfiles.

Generic AWS module patterns for serverless applications with Terraform. Use when organizing multi-environment deployments with tfvars, implementing naming conventions and tagging standards, configuring Aurora vs standalone PostgreSQL, setting up API Gateway with WAF, or building SNS/SQS event-driven architectures. Covers cross-account IAM role assumption, KMS encryption patterns, and module composition. Includes templates for naming helpers, tagging locals, and environment file organization.

GitLab CI/CD patterns for Terraform with AWS OIDC authentication. Use when setting up Terraform pipelines in GitLab, configuring HTTP backend for state storage, implementing OIDC-based AWS authentication (no static credentials), or managing multi-environment deployments. Covers pipeline stages (init, validate, plan, apply), resource groups for state locking, mise tool versioning, environment-specific jobs, manual approval gates, and MR-based plan previews. Includes templates for backend confi...

Testing strategy — unit/integration/E2E/performance planning with Jest, Vitest, Playwright, pytest. Coverage targets and CI/CD gates