The v2 rules inside one identical sandbox for every tool, with a provided headless Chromium for the agent’s own localhost preview. Runs are compared only within one protocol. How the protocols differ
One archived screenshot per published Run against version 1.3.0, of the Run’s representative Revision. Ordered by setup and then by repetition, not by quality.
Requirements
10 requirements · REQUIREMENTS.json
FJORA-01Campaign heroThe hero section shows the campaign name "Taste the North" and an agent-designed Fjora wordmark.
FJORA-02
Hero call to actionThe hero call-to-action scrolls or navigates to the flavor selector section.
FJORA-03Seasonal flavor optionsAll three seasonal flavors from products.json are listed as selectable options, each with its supplied product photo.
FJORA-04Flavor detail selectionSelecting a flavor updates the detail area to that flavor's supplied product photo, name, botanical list, and an agent-written tasting note; exactly one flavor is selected at all times and the first flavor is selected on load.
FJORA-05Detectable selection stateThe selected flavor option has a programmatically detectable selected state such as aria-pressed, aria-selected, or a radio group.
FJORA-06Brand storyThe story section presents an agent-written brand story with the supplied botanical macro image.
FJORA-07Expandable FAQAt least four agent-written FAQ entries render as expandable items, collapsed by default, individually expandable and collapsible via keyboard and pointer.
FJORA-08Fictional brand noticeThe footer contains the visible verbatim note "Fjora is a fictional brand created for UIBenchmark."
FJORA-09Mobile reachabilityAt 390px viewport width the page has no horizontal overflow and all required sections remain reachable.
FJORA-10Static production outputThe production build outputs a static site under dist/ that renders the page without a development server and without external network requests.
Frozen with this version. Requirement IDs are stable across versions and are cited unchanged by evaluation records.
Automated checks are not an accessibility certification. Recorded axe findings describe what the tooling observed at capture time, on a pinned browser and viewport. A check that reports no violation is not proof that the interface is accessible, and a selector problem is inconclusive rather than an implementation failure.
Run log
one fresh session per setup
Run log for version 1.3.0: one row per published Run
Operator-attested: the Configuration, tool version, displayed model, and elapsed time were recorded by the operator from what the tool showed at Run time. They are not independently verified. An exact model ID appears only when the tool exposed one; it is never inferred.
Published sites are independent, AI-generated demos served from separate preview origins with local mock data. They are not products, and no UIBenchmark session data crosses to them.
Execution records per Runshow +hide −
Exact model ID, tool version, elapsed time, run state, stop reason, discarded attempts and every published Revision, repairs included, in run log order. Each Run and each Revision links to its own page.
These Runs were executed against version 1.3.0, the most recently frozen Challenge Version under protocol v3. Runs against another frozen Version are listed on that Version’s own page, because they received different inputs.
Brief
Design and build Fjora's responsive Taste the North campaign landing page from the frozen prompt, rules, facts, photography, and requirements.
This is the brief of the most recently frozen Challenge Version under protocol v3. The full text of its launch prompt and shared rules, and the frozen inputs of earlier versions, are on the Challenge Version pages, so nothing here is a paraphrase of what a Run actually received.
Frozen versions (4)show +hide −
Newest first. A frozen version can never change; a correction becomes a new version.
Discarded attempts failed for a technical reason — a crashed session, a tool outage, a broken upload — and were discarded and retried from the same frozen inputs. The count is not a quality signal.
Generation cost: Not separately metered. Runs were executed on the operator’s personal tool subscriptions, which do not itemize a per-Run charge. No amount is attributed, and no zero-cost claim is made.
product-cloudberry.webp: license, source and SHA-256
webp
Path
public/assets/product-cloudberry.webp
License
Operator-generated for this Challenge
Redistribution
Yes
Source
AI-generated by the operator with GPT Images 2.5, September 2026
Provenance note
Fictional Fjora Cloudberry product photograph; no real-brand material.
