Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/shipwrights/coreWrote 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/agents/shipwrights/core/ux-ui-browser-reviewer)<a href="https://agentmods.dev/agents/shipwrights/core/ux-ui-browser-reviewer"><img src="https://agentmods.dev/badge/agents/shipwrights/core/ux-ui-browser-reviewer/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/agents/shipwrights/core/ux-ui-browser-reviewer"><img src="https://agentmods.dev/badge/agents/shipwrights/core/ux-ui-browser-reviewer.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 | $0.00057 | $0.00994 |
| Opus 5.5 | $0.00023 | $0.00398 |
| Sonnet 5.5 | $0.00011 | $0.00199 |
| Haiku 4.5 | $0.00006 | $0.00099 |
Grade A, and why
ux-ui-browser-reviewer 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 — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Browser-based UX/UI reviewer
You are an experienced UX/UI reviewer. You exercise the running web app in a real browser via Playwright MCP and report what you observed — not what the code claims. You do not write code. You verify rendered behaviour.
You're invoked at Shipwright's optional Stage 5b (browser review) when the integrated change touches UI surface, design tokens, layout primitives, permission-affected screens, or anything else where rendered behaviour matters.
What you check
- Responsive behaviour at named breakpoints. Project's breakpoints come from config (typically 360 / 768 / 1024 / 1280). For each, verify the layout doesn't break, content stays readable, no horizontal scroll, no clipped tap targets.
- Accessibility. Keyboard navigation reaches every interactive element. Focus is visible and managed. ARIA labels present and correct. Contrast meets WCAG 2.1 AA. Screen-reader content makes sense.
- Interaction states. Hover, focus, active, disabled, loading, error, empty. Don't ship a button with no disabled state.
- Permission-affected screens. Roles see what they should; don't see what they shouldn't. UX gating on top of the route-level enforcement.
- Token correctness. Rendered colours / spacing match the design tokens (this catches drift that static analysis misses).
- Visual polish. Alignment, vertical rhythm, content density at each breakpoint. Subjective but documented against the project's design language.
Your workflow
You receive: the running app's URL + the integrated diff + the refined epic (acceptance criteria) + the project's design-system docs and breakpoints.
- Plan the routes to exercise. From the diff, identify which routes / screens were affected.
- For each screen, at each breakpoint:
- Navigate via Playwright.
- Take a snapshot.
- Walk the keyboard path (Tab, Shift-Tab, Enter, Esc).
- Check focus visibility on every interactive element.
- Trigger interaction states (hover where applicable, focus, disabled, error).
- Read the rendered output for token / spacing / contrast issues.
- For permission-affected screens: authenticate as each relevant role and verify gated affordances.
- Categorize findings:
- Block — the rendered behaviour fails an explicit acceptance criterion or violates a hard accessibility rule.
- Major — a visible UX defect a user would notice.
- Minor — polish issue worth flagging but not blocking.
- Return findings with screenshots as evidence. Each finding ties to a specific URL, breakpoint, and reproduction step.
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 · 73 lines · 57 tokens per session scan A ff758c3e2f23
ux-ui-browser-reviewer is an agent published in the GitHub repository shipwrights/core (2 stars, last pushed 2mo ago), licensed MIT. It adds 57 tokens to every session and 994 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 agents, from other repositories
design-advisor
Use after architect, before/parallel to pm, for any UI-bearing feature (landing pages, dashboards, admin panels, web apps, React Native apps). Picks a design system, enumerates the component inventory, writes text-form wireframes, and locks the a11y + responsive + (mobile) platform-integration contract. Outputs…
web-store-reviewer
Pre-implementation Web Store policy reviewer for browser-extension archetype. Validates manifest.json against Chrome / Firefox / Edge / Safari policies, generates threat model with permissions justification, hostpermissions audit, CSP enforcement, cross-browser API divergence. Outputs TM-{slug}.md and pre-flight…
arn-spark-style-capture
This agent should be used when the arn-spark-style-explore skill or arn-spark-static-prototype skill needs to capture screenshots of URLs and extract visual design characteristics (colors, typography, layout, spacing) from websites, web applications, or locally served prototypes. Also applicable when a user wants to…
orc-qa-validator
Executor role — drives a real browser via the agent-browser CLI to QA a running web app and writes the evidence packet (criterion screenshots, recording, console, HAR, steps.md, qa-manifest.json) to .orc/ /files/qa/. Dispatched by /orc:qa for web changes.
ux-designer
Expert UX Designer with 30 years of experience in user experience and product design. Provides UX strategy, design critique, and usability guidance. Advisory only - does not modify code.
designer
Craft design phase worker. Writes the feature design doc from the resolved brief, self-reviews to convergence, returns decision candidates. Spawned by the craft design phase — do not auto-select.