Roman Senchuk
Community@r-senchuk · Poland
Roman Senchuk publishes MIT-licensed skills for Next.js static-export sites, Tailwind CSS v4 interfaces, LLM agent engineering, and macOS shell environments.
Agent Skills by Roman Senchuk
Showing 23 vetted skills indexed across 1 GitHub repositories.
before-after-slider
Implement accessible before/after image comparison sliders with drag, touch, and keyboard support in Next.js.
agent-testing
Builds tiered test suites for AI agents with mocked unit tests, integration tests, and eval frameworks.
photo-upload-form-ux
Implements quote request funnels with photo upload guidance for static Next.js sites.
nextjs-ssg
Scaffold, configure, and debug Next.js 16 App Router projects with static export.
shell-script-audit
Audit and harden bash and zsh scripts for error handling, quoting, portability, and security.
sara
Route repository tasks to specialist agents and review delegated results.
cro-home-services
Designs conversion-focused UX patterns for home services renovation websites.
typescript-7
Migrate Node.js and Next.js projects to the native TypeScript 7 compiler.
mobile-first-layout
Implements responsive mobile-first page layouts and grids for Tailwind CSS v4 Next.js sites.
skill-builder
Create, audit, and refactor SKILL.md files against Agent Skills quality standards.
macos-homebrew-troubleshoot
Diagnose and fix macOS environment, Homebrew, PATH, and launchd service issues.
agent-builder
Create and validate .agent.md files with scoped personas, minimal tools, and constraints.
trust-signal-components
Implement trust signal UI components for home services renovation websites.
nextjs-tailwind-seo
Configure Tailwind CSS 4.3, SEO metadata, fonts, and structured data in Next.js 16 App Router projects.
imagery-art-direction
Select, optimize, and place WebP photography for static Next.js renovation websites.
context-engineering
Design token-budgeted context assembly, memory, and tool-result injection for LLM agents.
harness-engineering
Design guardrails, feedback loops, state persistence, and error recovery for production AI agent systems.
nextjs-intl
Configure next-intl v4 internationalization for Next.js 16 App Router projects with static export.
visual-design-audit
Audit pages and components against brand tokens, accessibility, and CRO patterns to generate prioritized findings.
edd-loop
Executes evaluation-driven implementation loops with bounded workers, deterministic gates, and fresh read-only verifiers.
tailwind-v4-theming
Add, extract, and debug Tailwind CSS v4 design tokens using CSS-first @theme configuration.
floating-sticky-ui
Implement and debug sticky headers, floating CTA bars, and z-index layering in Tailwind Next.js sites.
zsh-config-expert
Configure and troubleshoot zsh startup files, completion systems, prompts, and plugin frameworks.
Frequently Asked Questions About Roman Senchuk
FAQPage SchemaWhat tasks can I accomplish with Roman Senchuk's skills?▼
You can scaffold Next.js 16 static-export sites, configure Tailwind CSS v4 design tokens, add next-intl internationalization and SEO metadata, build conversion-focused renovation site components, test LLM agents with promptfoo and DeepEval, design agent context and harness layers, and audit bash/zsh scripts or macOS Homebrew environments.
Who are these skills designed for?▼
Front-end engineers building static Next.js marketing sites (especially home services businesses), agent engineers working with LangChain, CrewAI, or OpenAI Agents SDK who need testing and context engineering, and developers maintaining zsh, bash, and Homebrew setups on macOS.
What are the runtime requirements for the Next.js skills?▼
The Next.js skills require Node.js 20.9+, Next.js 16.2.11+ (Active LTS), TypeScript 7.0+, and pnpm. UI component skills target Next.js 15+ with Tailwind CSS v4 and React 18+ on static export sites. The agent-testing skill requires Python 3.9+ with pytest and pytest-asyncio.
Are these skills free and open source?▼
Yes. Nearly all skills carry an MIT license, permitting free use, modification, and redistribution. A few skills (shell-script-audit, macos-homebrew-troubleshoot, agent-builder, context-engineering, harness-engineering, edd-loop) list no explicit license in their frontmatter.
How do I invoke the skills in practice?▼
Most UI and funnel skills are user-invocable with an argument hint such as image paths, locale lists, or target sections. Orchestration skills like sara and edd-loop require explicit invocation (/sara or $edd-loop) and disable automatic model triggering, while audit skills activate via file paths like next.config.* or tsconfig.json.