Menu
Use when scanning, auditing, comparing, or testing UI with IBR.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
Use when /ibr:build detects .mockup-gallery/ or IBR needs to tie a build to gallery-selected mockups. Reads ratings/selections, enforces approved target roles, preserves scratch-first workflow, and writes implementation completion. Filesystem only.
Use when validating iOS, watchOS, or macOS UI with IBR.
Use when working on iOS, watchOS, or macOS apps — .swift files, SwiftUI views, simulators. IBR's native scan workflow for touch targets, a11y labels, watchOS constraints, Fix Guide.
Use when planning or implementing UI with IBR.
Use when planning, designing, implementing, or auditing UI/UX with IBR in Codex. Provides the latest compact IBR guidance for Calm Precision, web archetypes, interaction states, mobile, content states, data visualization, Mockup Gallery targets, and imagegen approval gates.
Use when building cards, navs, forms, dashboards, modals, tables, or lists. Opinionated patterns with Calm Precision principles, spacing, accessibility, and anti-patterns.
| name | validate |
| description | Use when scanning, auditing, comparing, or testing UI with IBR. |
Use IBR MCP tools as evidence, not decoration. Validate the rendered interface against user intent, design intent, accessibility, interaction behavior, and console health.
scan: full web page scan for layout, styles, semantic state, accessibility, handlers, and console issues.snapshot: capture a before state before risky UI edits.compare: verify whether current UI changes are expected or regressions.screenshot: capture visual evidence when layout, canvas, media, or design-match judgment needs pixels.observe: list actionable elements by accessible role and name.interact or interact_and_verify: click, type, select, and verify state changes. success on the response reflects a real expected-outcome validator, not just that the call didn't throw — a no-op click (target resolved, nothing changed) returns success: false with validator/evidence explaining what was expected vs. observed. Read validator.passed, don't assume success is always true.flow_search, flow_form, flow_login: validate common task flows.match: compare an approved visual target against a live page..ibr/builds/<topic>/design-intent.json.Report the actual issue and the file or UI area it affects. Do not treat an ISSUES verdict as acceptable without reading the issue list. If the tool cannot run, state the exact blocker and the fallback evidence used.