Design intelligence + stack-aware skills for Claude Code. Auto-detects your stack and routes to the right skill. Companion: superpowers@claude-plugins-official.
Use for systematic codebase audits and large-scale refactoring. Conducts a full four-category analysis (security, performance, architecture, directory structure) before making any changes, then presents findings with severity ratings and a fix plan. Best triggered when the user says "analyze the project", "audit the codebase", "find problems", or "refactor the project". Examples: <example>user: "See this project, analyze it thoroughly and report security loopholes, performance bottlenecks, architecture flaws, and directory structure flaws" assistant: "I'll use the audit-refactor agent to conduct a full four-category analysis."</example> <example>user: "The codebase has gotten messy, do a full audit" assistant: "Dispatching audit-refactor agent for a structured analysis before any changes."</example>
Use for a post-generation quality pass before PR review, or after AI-assisted coding sessions. Applies a comprehensive checklist covering TypeScript, architecture, naming, slop removal, inline type extraction, and import hygiene. Examples: <example>user: "Clean up the code I just wrote for the dashboard feature" assistant: "I'll use the code-cleanup agent to apply full quality standards to the recent changes."</example> <example>user: "Review this before I push" assistant: "Running code-cleanup agent across the changed files."</example>
Use when starting a new feature, component, page, or API endpoint in a Next.js project. Implements via strict data-flow sequence with version-aware docs fetching. Examples: <example>user: "Build the user profile settings page" assistant: "I'll use the feature-builder agent to implement this following the correct data flow."</example> <example>user: "Add a create order endpoint with optimistic updates" assistant: "Dispatching feature-builder — will fetch current Next.js/React 19 docs, then implement from types through to UI."</example>
Use to remove AI-generated artifacts from code: redundant comments, type escape hatches, abnormal defensive checks, debug logging, inline complex types, process journals, and all other patterns inconsistent with how a careful human developer writes code. Examples: <example>user: "Remove all the AI slop from this PR" assistant: "I'll use the slop-remover agent to systematically identify and remove all AI-generated artifacts."</example> <example>user: "The code has too many comments explaining obvious things" assistant: "Dispatching slop-remover — will scan for all slop patterns and remove them."</example> <example>user: "Remove AI code slop" assistant: "Using slop-remover to clean up the codebase."</example>
Use when writing browser automation, E2E tests, or interaction validation with Playwright; when the user asks to test a UI flow, automate a browser action, check responsive layouts, validate form submissions, or verify link integrity. Requires Playwright installed in the project: npm install -D @playwright/test && npx playwright install.
Use when creating a new UI component, page, screen, or layout from scratch; building a landing page, dashboard, SaaS UI, portfolio, or any greenfield frontend interface. Do NOT use for auditing or polishing existing UI — use ui-polish for that. After completing UI generation, always invoke ui-polish for a quality pass.
Use when reviewing, auditing, critiquing, polishing, or improving existing UI; checking a component for design issues; iterating on styling quality. Also runs automatically as the second step of the ui-new pipeline after generating new UI. Commands: audit, polish, critique, shape, craft, animate, harden, and 16 more.
Use when creating interactive prototypes, HTML slide decks, motion graphics, GIF or MP4 animations, or infographics; when presenting design directions or comparing design variations; when the user wants to visualize ideas without writing production code. Do NOT use for standard component or page development — use ui-new for that.
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Design intelligence + stack-aware skills for Claude Code. Auto-detects your project stack and routes to the right skill — no configuration needed.
| Component | Source | Purpose |
|---|---|---|
polaris:ui-new skill | UI UX Pro Max | Generate new UI from scratch with industry-specific design systems |
polaris:ui-polish skill | Impeccable | Audit, refine, and quality-check UI with 27 anti-pattern rules |
polaris:ui-prototype skill | Huashu Design | Create prototypes, slide decks, motion graphics, and infographics |
polaris:playwright-e2e skill | playwright-skill | Browser automation and E2E testing with Playwright |
/init command | Karpathy 12-rule template | Creates or audits CLAUDE.md with behavioral rules + project context |
| Frontend baseline rules | Taste-Skill | Always-on design quality constraints (no AI purple, no emoji icons, etc.) |
| Next.js agents | abhisheksharm-3 gist | Four specialized agents for Next.js 15+ / React 19 workflows |
| Auto-detect hook | Custom | Detects stack at session start, injects the right rules automatically |
Polaris handles design and stack workflows. For process skills (TDD, debugging, planning, code review), install Superpowers alongside it:
/plugin install superpowers@claude-plugins-official
/plugin marketplace add abhisheksharm-3/polaris
/plugin install polaris
claude --plugin-dir ./polaris
Polaris auto-injects rules and agent routing for:
"next" in package.json)"react" in package.json, no Next)pyproject.toml or requirements.txt)Cargo.toml)go.mod)"playwright" or "@playwright/test" in package.json)/init — CLAUDE.md generatorRun /init (or /polaris:init) at the root of any project to set up a CLAUDE.md file.
No existing CLAUDE.md: scans your project (commands, architecture, README, cursor rules) and generates a file combining the Karpathy 12-rule behavioral template with your project-specific context.
CLAUDE.md already exists: audits it for missing behavioral rules and stale project sections, then proposes targeted diffs.
The generated file always includes all 12 behavioral rules and stays under the 200-line compliance ceiling.
| You want to... | Skill or agent used |
|---|---|
| Build new UI from scratch | polaris:ui-new → polaris:ui-polish |
| Audit or refine existing UI | polaris:ui-polish |
| Create a prototype or slide deck | polaris:ui-prototype |
| Write E2E tests | polaris:playwright-e2e |
| Build a Next.js feature | polaris:feature-builder agent |
| Post-AI-session cleanup | polaris:code-cleanup agent |
| Full codebase audit | polaris:audit-refactor agent |
| Remove AI-generated artifacts | polaris:slop-remover agent |
All skill content is MIT licensed. Source repositories:
MIT — see LICENSE file.
npx claudepluginhub abhisheksharm-3/polaris --plugin polarisComprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Comprehensive startup business analysis with market sizing (TAM/SAM/SOM), financial modeling, team planning, and strategic research
v9.44.1 — Patch release for Gemini environment/version detection and qwen auth gating. Run /octo:setup.
Complete creative writing suite with 10 specialized agents covering the full writing process: research gathering, character development, story architecture, world-building, dialogue coaching, editing/review, outlining, content strategy, believability auditing, and prose style/voice analysis. Includes genre-specific guides, templates, and quality checklists.
Comprehensive .NET development skills for modern C#, ASP.NET, MAUI, Blazor, Aspire, EF Core, Native AOT, testing, security, performance optimization, CI/CD, and cloud-native applications