AI capability boundaries, agent control, source grounding, uncertainty, approval, and safe fallback.
Open in lab Source-backed UX patterns 298 documented entries 897 sources
Choose the right UX pattern before the interface hardens.
A practical decision system for product teams, designers, frontend engineers, and AI agents: search the problem, compare nearby patterns, inspect required states, and verify the source trail.
- 298
- pattern entries
- 298
- documented patterns
- 286
- comparisons
- 897
- evidence sources
Use it like a product decision system
The site is structured around the work a real team has to do before an interaction ships.
Review Use the checklist to define task fit, risk, required states, accessibility checks, and evidence before the choice becomes a ticket. Choose Start from the user problem, compare nearby alternatives, and avoid choosing a component by name alone. Inspect Open the lab to test states, keyboard behavior, failure modes, and implementation contracts before copying a pattern. Prove Trace guidance back to sources, verification dates, and source-backed claims when a pattern affects risk or trust.
Pattern families
Explore all fourteen families when you know the broader problem area.
Browse all pattern families 14 families
Comments, mentions, presence, feeds, handoffs, shared attention, and team workflow.
Open in labTouch, haptics, sensors, camera, permissions, offline states, and device adaptation.
Open in labTables, dashboards, charts, maps, compare views, timelines, and drill-down behavior.
Open in labProgressive disclosure, overlays, warnings, focus, and interruption management.
Open in labValidation, undo, retry, fallback, destructive action review, and recovery paths.
Open in labLoading, progress, alerts, banners, sync, confidence, and system visibility.
Open in labForms, field behavior, uploads, dates, addresses, payment entry, and account data.
Open in labHeaders, breadcrumbs, tabs, side navigation, steps, local anchors, and return paths.
Open in labSettings, preference centers, saved views, defaults, recommendations, and user control.
Open in labSearch, filters, facets, category browsing, query correction, and no-results recovery.
Open in labChoice controls, object pickers, ranges, menu actions, bulk selection, and commands.
Open in labMulti-step work, approvals, scheduling, review queues, checkout, and task completion.
Open in labConsent, privacy, security, sensitive data, reporting, permissions, and safety exits.
Open in labHigh-stakes patterns to inspect first
These entries show the standard every page is moving toward: decision rule, quality bar, implementation contract, and evidence trail.
AI limitation onboarding
A user is introduced to an AI feature whose abilities, limits, data scope, uncertainty, or review needs are not obvious from the normal interface.
AI And Automation UX - 5 sourcesEmpty state
The product area can legitimately contain no user data.
Feedback, Status, And System State - 4 sourcesError state
A system or task failure blocks expected content or action.
Feedback, Status, And System State - 4 sourcesHuman approval gate
An AI agent, workflow, deployment, or automation is ready to perform a high-impact step and must pause for human authorization.
AI And Automation UX - 6 sourcesInline validation
A single field has a specific correctable problem.
Input And Data Entry - 2 sourcesSource grounding display
Users need answer-wide evidence coverage before trusting generated content.
AI And Automation UX - 7 sources