Borrowing it
Nothing to install: this file belongs to sourjya/kiro-rails. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/sourjya/kiro-rails/main/.claude/commands/review-ux-live.mdgit clone --depth 1 https://github.com/sourjya/kiro-railsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/sourjya/kiro-rails/review-ux-live)<a href="https://agentmods.dev/commands/sourjya/kiro-rails/review-ux-live"><img src="https://agentmods.dev/badge/commands/sourjya/kiro-rails/review-ux-live/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/sourjya/kiro-rails/review-ux-live"><img src="https://agentmods.dev/badge/commands/sourjya/kiro-rails/review-ux-live.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00055 | $0.01962 |
| Opus 5.5 | $0.00022 | $0.00785 |
| Sonnet 5.5 | $0.00011 | $0.00392 |
| Haiku 4.5 | $0.00006 | $0.00196 |
Grade A, and why
review-ux-live scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 220 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Live UX Review - Browser-Walk Protocol
A structured protocol for reviewing a running application's UX via browser
automation. Each page is walked with 9 steps that produce evidence for
scoring against the "UX Console-Idiom Rubric" rules in CLAUDE.md.
Prerequisites:
- A browser MCP (Playwright MCP, Chrome DevTools MCP, or BrowserTools MCP)
- The application running and accessible (local or deployed)
scripts/style-survey.jsavailable for injection
Before you start: Load the "UX Console-Idiom Rubric" rules in CLAUDE.md for the
rubric definitions and scoring model.
Side-Effect Boundary - NON-NEGOTIABLE
During a live review, you are an observer only. Do NOT:
- Submit forms (except safe dirty-state probes - see Step 5)
- Click delete, remove, or destructive action buttons
- Send messages, emails, or notifications
- Log out (you may lose session state)
- Create, modify, or save data
- Accept or dismiss important system dialogs that affect state
Safe interactions ONLY:
- Navigation (click links, tabs, breadcrumbs)
- Hover and focus (for tooltip/state discovery)
- Type into fields WITHOUT submitting (dirty-state probe)
- Press Escape (overlay dismissal test)
- Scroll (full-page capture)
- Open dropdowns and menus (read-only discovery)
If unsure whether an action is safe, do not take it. Report it as "unverifiable without write access" in the findings.
Per-Page Protocol (9 Steps)
Run these 9 steps for each page or major route in the application.
Step 1: Verify Session State
Before interacting, confirm:
- URL is correct (not redirected unexpectedly)
- Authentication state is valid (user is logged in if required)
- The page has fully loaded (no spinners, no partial content)
If the page redirected or shows an error, document it as a finding before proceeding.
Step 2: Screenshot + Accessibility Tree
Capture two things simultaneously:
- A full viewport screenshot (evidence for visual findings)
- The accessibility tree or ARIA snapshot (evidence for A-family checks)
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 220 lines · 55 tokens per session scan A fb3a760509a6
review-ux-live is a command published in the GitHub repository sourjya/kiro-rails (9 stars, last pushed 3mo ago), licensed MIT. It adds 55 tokens to every session and 1,962 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
naksha-browse
Browse any URL with browser vision — extract design patterns for research or inspect your own UI, saved to project memory.
design-compare
Capture two websites with Playwright and produce a structured side-by-side design comparison — layout, color, typography, navigation, and 'steal this' recommendations.
live-canvas
Conduct design interviews, generate UI variations, and collect live click-to-annotate feedback that streams into the session so edits land without leaving the browser. Use when the user wants rapid iterative UI refinement, not just batched feedback.
ui-ux-verify
Read-only verification of one rendered page against one OpenSpec UI specification. Use when: checking a page against a spec, auditing a controller/action render, or reviewing UI content, structure, and behavior. Not for: E2E journey authoring, application edits, data mutation, or a runtime API smoke test. Output: one…
annotate
Add visual annotations (boxes, highlights, callouts) to the current page and take a screenshot. Use during user testing to highlight issues or confusion points.
live-canvas
Conduct design interviews, generate UI variations, and collect live click-to-annotate feedback that streams into the session so edits land without leaving the browser. Use when the user wants rapid iterative UI refinement, not just batched feedback.