Skip to main content
rule CX Codex

review-and-research — Claude Code rule: what it does, how to install

review-and-research standardizes UI audits, usability checks, and design handoff requirements. Activates when Claude Code inspects screens or prepares components for developers.

MIT ★ 3,216 updated 2026-10-20

Install into Claude Code · Plugin marketplace All methods and platforms ↓
/plugin marketplace add plugin87/ux-ui-agent-skills
/plugin install ux-ui-agent-skills@ux-ui-agent-skills
Open on GitHub

What review-and-research does

This rule guides Claude Code through interface reviews using six weighted dimensions. Usability and hierarchy carry the most weight. Every finding receives a severity label, alongside checks against Nielsen's 10 heuristics.

For interface planning, it introduces a five-stage fidelity ladder from plain text to running code. The handoff stage enforces token mapping, accessibility attributes, and documentation for eight element states. Without files in the workflows directory, the agent misses full rubric definitions.

What it helps with

  • audit an interface against Nielsen's 10 usability heuristics
  • grade design quality across six weighted criteria
  • verify UI components before developer handoff
  • plan interface validation across five fidelity stages

Example prompts for the agent

  • Run a design audit on this view and list findings by severity.
  • Check this modal component against the design-to-code handoff checklist.
  • Review our wireframe using Nielsen's usability heuristics.

How to install review-and-research

1Plugin marketplace

/plugin marketplace add plugin87/ux-ui-agent-skills
/plugin install ux-ui-agent-skills@ux-ui-agent-skills

2Manually: put the files in place

.claude/rules/review-and-research.md

The paths frontmatter field limits the files

Step by step, with commands for each platform: AGENTS.md vs CLAUDE.md: what they are and which one to use

Questions about review-and-research

How to install review-and-research in Claude Code?

Run /plugin marketplace add plugin87/ux-ui-agent-skills, then run /plugin install ux-ui-agent-skills@ux-ui-agent-skills.

What does the review-and-research rule do?

It instructs Claude Code to score design quality across six metrics, apply Nielsen's heuristics, and verify that components meet handoff standards.

What is required before marking a design ready for development?

Designs must map to tokens without hardcoded values, define eight interactive states, handle responsive breakpoints, and include accessibility notes.

Source text

review-and-research.md · MIT
# Rule: Review, research, and handoff

Loaded when reviewing or auditing a design, prototyping, or handing off to development.

Split out of `CLAUDE.md` so it loads only when the work calls for it. The
headings below are unchanged, so any pointer to them still resolves.

---

## Design Review & Audit

### When to Use
- **Design Review**: Evaluating a new design before development
- **Design Audit**: Evaluating an existing product for consistency and quality

### Review Output Format
Score across 6 dimensions (1 to 10), then provide a prioritized findings table:

| Dimension | Weight | Score |
|-----------|--------|-------|
| Visual Hierarchy | 20% | ?/10 |
| Consistency | 20% | ?/10 |
| Accessibility | 20% | ?/10 |
| Usability | 20% | ?/10 |
| Responsiveness | 10% | ?/10 |
| Performance | 10% | ?/10 |
| **Overall** | **100%** | **?/10** |

Then findings:

| # | Severity | Finding | Recommendation |
|---|----------|---------|---------------|
| 1 | Critical | [what's wrong] | [how to fix] |
| 2 | Major | ... | ... |

Severity levels: **Critical** (must fix before launch) → **Major** (fix this sprint) → **Minor** (fix when convenient) → **Enhancement** (backlog)

Full rubric and process: `workflows/design-review.md`

### Heuristic Evaluation
Apply Nielsen's 10 Usability Heuristics to every review. Flag violations with the heuristic number. See `workflows/design-review.md` for the full checklist.

---

## Prototyping & Research

### Fidelity Ladder
Never skip fidelity levels. Validate at each stage:

| Level | Output | Time | Validate |
|-------|--------|------|----------|
| 1. Content-first | Text outline | 30 min | Information needs |
| 2. Wireframe | Box layouts | 1-2 hr | Layout & navigation |
| 3. Low-fi prototype | Clickable flows | 2-4 hr | Task completion |
| 4. High-fi mockup | Pixel-perfect | 4-8 hr | Visual & a11y |
| 5. Code prototype | Working code | 1-3 days | Feasibility & performance |

### User Research Methods
- Card sorting → navigation structure
- Tree testing → findability validation
- Usability testing → task success rates (5 users catches 85% of issues)
- See full methodology and scripts in `workflows/prototyping.md`

---

## Design-to-Code Handoff

### Handoff Checklist
Before marking a design ready for development:
1. All values mapped to design tokens (zero hardcoded values)
2. All 8 states documented per interactive element
3. Edge cases addressed (long text, empty, overflow, single item, many items)
4. Responsive behavior spec'd at each breakpoint
5. Animation spec'd (property, duration, easing, reduced-motion fallback)
6. Accessibility annotations (ARIA roles, keyboard model, focus management)

### Definition of Done
A component is done when: functional (all variants, states, edge cases), visual (pixel-accurate, all tokens, responsive, dark mode), accessible (keyboard, screen reader, contrast, target size), code quality (TypeScript, no `any`, forwardRef, cva), and tested (unit, visual regression, a11y automated, manual screen reader).

Full workflow: `workflows/design-to-code.md`

© Thientan Soparat · MIT · plugin87/ux-ui-agent-skills

Similar IDE rules

All IDE rules for Design →

From the same set35

The whole set →
observe observe — a skill for Claude Code that checks local hook events to see if the agent followed UX/UI kit rules in recent sessions. Runs manually on user request. CX Codex +1 ★ 3,216repo stars frameworks frameworks directs Claude Code to generate production UI code for React with Tailwind v4, Next.js 15, and SwiftUI 6 without placeholders. CX Codex +1 ★ 3,216repo stars ship ship — a skill for Claude Code runs pre-release checks and drafts a release checklist. It starts before a release tag, checking the project before preparing version and changelog details. CX Codex +1 ★ 3,216repo stars gate Skill for Claude Code runs a UI quality gate before marking tasks complete. You trigger the command manually, preventing the model from calling it autonomously. CX Codex +1 ★ 3,216repo stars brand-and-operations brand-and-operations is a Claude Code rule for brand consistency, motion, and UI copy. It loads when an agent sets an aesthetic direction, writes interface text, or works on system operations. CX Codex +1 ★ 3,216repo stars typography-and-spacing typography-and-spacing is a Claude Code rule for choosing type scales and layout spacing when an agent works on page typography or rhythm. It also covers checks for Thai text. CX Codex +1 ★ 3,216repo stars scaffold-project scaffold-project — a Claude Code skill that scaffolds a product design workspace with design tokens, rules, and folder structure. Runs when initiating a fresh design system or web application. CX Codex +1 ★ 3,216repo stars critique The critique Claude Code skill challenges a finished interface and finds grounds to reject it. Run it after the project gates pass. CX Codex +1 ★ 3,216repo stars

Project sponsors

Companies that keep this analytics open