Manifest position 628 · CrUX rank bucket 1000
https://www.emag.ro
Coverage complete
Atomic coverage complete across seven representative public paths; checkout payment and authenticated account internals were intentionally not transacted.
Attempts
1 / 3
Judged checks
58 / 58
Blocked
0
Not run
0
This report has no published overall score. Blocked and not-run checks are not passes. A coverage-complete report means every check has a judged outcome; it does not mean every check passed.
All 58 atomic check outcomes
| Principle / check | Status | Confidence | Evidence or reason |
|---|---|---|---|
respect-user-preferencesrespects-color-scheme | issues | high | The dark-mode screenshot is visually identical to the light capture; the preference probe reports body background rgb(245,245,245), color-scheme normal, and no prefers-color-scheme rules. F01 medium: The site stays light when the user requests dark mode. |
respect-user-preferencesrespects-reduced-motion | pass | high | Under prefers-reduced-motion: reduce, the probe reports zero active Web Animations and the repeated carousel probe still reports zero animations. |
respect-user-preferencesrespects-contrast | pass | high | Forced-colors/prefers-contrast capture keeps navigation, search, consent actions, carousel controls, and text visibly distinguishable. |
implement-natural-interactionsview-transitions | issues | high | Rendered CSS/API marker inspection found zero view-transition rules, while navigation uses conventional cross-document swaps. F02 low: Page and state changes do not use View Transitions. |
implement-natural-interactionsscroll-driven-animations | not-applicable | high | No scroll-linked parallax, scrollytelling, or reveal interaction was observed on the representative routes, so there is no scroll-linked effect to implement. |
implement-natural-interactionsphysical-gestures | not-applicable | high | No custom gesture-driven removal, pull-to-refresh, or comparable pointer interaction was exposed in the audited public flow; ordinary document scrolling remains platform-native. |
provide-guided-navigationscroll-state-aware-chrome | issues | high | The 5,797px homepage and 17,898px product page have static multi-row headers; source markers show no scroll-state or scroll-driven navigation technique. F03 low: Long pages do not expose scroll-state-aware chrome. |
provide-guided-navigationanchored-positioning | issues | high | The load screenshot shows an account tooltip attached to the header icon, but rendered CSS contains no anchor-name, position-anchor, or position-try markers. F04 low: The account prompt is custom-positioned rather than anchor-positioned. |
provide-guided-navigationdirects-attention | pass | high | Product breadcrumbs, descriptive route headings, selected carousel dots, and visible focus treatment give location and focus cues on tested routes. |
maximize-content-reduce-noiseno-intrusive-interruptions | issues | high | Desktop load shows a large cookie banner covering the hero while an account sign-in prompt covers the navigation and hero; mobile shows the cookie panel covering product cards. F05 high: Two simultaneous load-time interruptions obscure the primary content. |
maximize-content-reduce-noisesemantic-dismissible-primitives | issues | high | The homepage probe counted zero dialog, popover, and details elements despite visible cookie and account overlays. F06 medium: Visible overlays are built without semantic dialog or popover primitives. |
maximize-content-reduce-noisereduced-chrome | issues | high | The desktop first viewport contains an app-install strip, two navigation rows, an account prompt, a hero promotion, cookie controls, and an accessibility widget; little unobstructed shopping content remains. F07 medium: Promotional chrome crowds the first viewport. |
adapt-to-the-form-factorresponsive-no-horizontal-scroll | issues | high | At 360px, layout reports 840px horizontal overflow on the category route (1,200px layout viewport) and 27px on the product route; the homepage itself has none. F08 high: Representative catalogue and product pages overflow on mobile. |
adapt-to-the-form-factorcomponent-level-responsiveness | issues | high | The category route forces a 1,200px layout at a 360px visual viewport, and rendered styles expose no @container rule. F09 medium: Shared catalogue components do not adapt to their available container. |
adapt-to-the-form-factorinput-modality-aware | pass | high | The first keyboard-focusable control receives a visible 5px auto outline, and Lighthouse did not report target-size failures. |
support-core-task-successclear-purpose-and-primary-action | pass | high | Search is prominent on home, the product route names the exact item and shows price/rating, and “Adauga in Cos” controls are present. |
support-core-task-successprimary-flow-completion | pass | high | The product route exposes multiple add-to-cart controls and the cart route loads an explicit cart surface with recommendations rather than an account wall. |
support-core-task-successclear-system-state-and-recovery | issues | high | The deliberately invalid route renders a mostly blank carousel shell with arrows and dots, no 404 message, no search prompt, and no route recovery action. F10 high: The not-found route does not explain the failure or offer recovery. |
be-fast-and-stablegood-core-web-vitals | issues | high | Lighthouse performance is 0.62 with FCP 4.4s and LCP 10.3s; an unthrottled trace is faster at LCP 1.36s, showing network/CPU conditions are the risk rather than instability. F11 high: Throttled mobile loading misses the LCP target. |
be-fast-and-stablevisual-stability | pass | high | Mobile homepage CLS is 0.0030, product CLS is 0, and Lighthouse CLS is 0.002 in the observed load windows. |
be-fast-and-stableefficient-main-thread | pass | high | Lighthouse TBT is 150ms; the trace records 107.57ms TBT with two long tasks and the mobile layout observer records one 111ms task, remaining within a usable initial-load range. |
be-fast-and-stableefficient-resource-delivery | issues | high | HAR records 82 requests and 1,847,353 transferred bytes; Lighthouse flags LCP discovery and network dependency tree insights. F12 medium: Critical delivery is heavier and less direct than necessary. |
be-fast-and-stabletrim-unused-and-duplicate-code | issues | high | Lighthouse estimates 288,526 bytes of unused JavaScript and 63,583 bytes of unused CSS; HAR attributes 653,632 transferred bytes to scripts. F13 medium: The initial route ships substantial unused code. |
be-inclusivenames-roles-labels | issues | high | Lighthouse fails label-content-name-mismatch; the homepage probe estimates 34 unnamed interactive elements and the login email input has no associated label or aria-label. F14 medium: Some controls and fields lack robust accessible naming. |
be-inclusivesufficient-contrast | issues | high | Lighthouse reports three color-contrast failures on the homepage. Forced-colors remains usable, but that does not repair default-theme WCAG contrast. F15 medium: Several text/UI combinations fail minimum contrast. |
be-inclusivestructure-and-focus | issues | high | The homepage probe finds no h1 and no header landmark; content begins with many h2 product titles. Visible focus does work on the tested link. F16 medium: The homepage heading/landmark structure starts without a page-level heading. |
be-inclusivelegible-text | pass | high | Desktop/mobile screenshots show readable body and heading text with normal wrapping; no text clipping was observed on the responsive homepage. |
be-inclusivezoom-reflow-targets-and-media | issues | high | The 360px category and product tests report horizontal overflow of 840px and 27px respectively despite a scalable viewport meta tag. F17 high: Mobile reflow fails on key catalogue routes. |
follow-best-practicesno-console-errors | pass | high | Lighthouse errors-in-console audit scores 1 and its best-practices category scores 1. |
follow-best-practicessound-document-and-assets | issues | high | The image audit found 51 of 54 images without width/height, 19 oversized, and 46 legacy-format images. CLS happened to remain low in this capture but reservation is not guaranteed. F18 medium: Image markup omits dimensions and often delivers unsuitable assets. |
follow-best-practicesbrowser-platform-hygiene | issues | high | Lighthouse reports three BFCache failure reasons, including cache-control:no-store on the main resource and a JavaScript request. F19 medium: The homepage cannot use the back/forward cache. |
be-discoverabletitle-and-description | pass | high | Home, category, and product probes each expose a route-specific title and non-empty meta description. |
be-discoverablecrawlable-and-mobile-friendly | issues | high | Lighthouse fails crawlable-anchors; probes find 25 javascript/no-href links on home and 7 javascript links on category. The category route also forces a 1,200px layout at 360px. F20 high: Some links are not crawlable and one public archetype is not mobile-friendly. |
be-discoverablecanonical-and-indexing-signals | issues | high | Direct requests to /robots.txt and /sitemap.xml both returned HTTP 410 with the same 587-byte generic Gone page, even though canonical tags and public pages return 200. F21 high: Robots and sitemap endpoints return 410 Gone. |
be-discoverablestructured-and-shareable-metadata | pass | high | The product route exposes Product/Breadcrumb/QAPage JSON-LD plus matching Open Graph title, description, image, URL, and product type. |
be-private-and-securesecure-transport-and-headers | issues | high | The page is HTTPS with HSTS, but all 14 observed first-party cookies lack Secure and use SameSite=None; CSP allows unsafe-inline and unsafe-eval. F22 high: Cookie transport attributes and CSP are weaker than the HTTPS transport. |
be-private-and-securedata-minimisation-and-third-parties | issues | high | The tracker capture sees 9 third-party origins, including Google Tag Manager and Sentry, while the cookie decision panel is still visible in the load screenshots. F23 medium: Tracking and third-party activity starts on the initial page load. |
be-private-and-securein-context-permissions-and-modern-auth | issues | high | The login probe finds only email plus Facebook/Google/Apple controls and zero passkey/WebAuthn markers; no permission prompt fired on load. F24 medium: The relevant sign-in route exposes no phishing-resistant sign-in option. |
be-private-and-securedefensive-browser-policies | issues | high | Permissions-Policy is missing, Referrer-Policy is the legacy no-referrer-when-downgrade value, and CSP is wildcard-based with unsafe-inline/unsafe-eval, although HSTS, nosniff, SAMEORIGIN and frame-ancestors are present. F25 medium: Defensive browser policy is incomplete. |
be-resilientprogressive-enhancement | pass | high | Discoverability captures show server-rendered content rather than JS shells: 74% raw/rendered token coverage on category and 88% on product, with title, h1 and description present without JS. |
be-resilientresilient-runtime-behaviour | pass | high | Observed account and consent overlays remain within the viewport at desktop and forced-colors conditions, and core product/category content still renders around them. |
be-resilientoffline-and-installable | issues | high | The homepage promotes a mobile app, but the PWA probe finds no manifest and no service-worker registration. F26 medium: The app-like commerce experience has no install or offline foundation. |
be-resilientnetwork-and-http-failure-states | issues | high | The invalid-route capture shows only empty carousel chrome and no useful status or recovery controls. F27 high: Failure handling can collapse into an unexplained shell. |
be-internationalisedlang-dir-and-logical-properties | pass | high | Representative probes report html lang=ro; Romanian is left-to-right so omitted dir correctly defaults to ltr, and visible reading order is coherent. |
be-internationalisedlocale-aware-data | pass | high | The product page renders Romanian-style currency and decimals, including “6.699,99 Lei” and “4.73”, with Romanian localized copy. |
be-internationalisedtime-zone-correctness | not-applicable | high | The audited catalogue, product, cart, and login surfaces expose no scheduled event or user-entered time-zone-sensitive data to test. |
be-trustworthyno-dark-patterns | pass | high | The visible consent panel gives Accept and Refuse equal-sized, equally prominent controls plus preference management; product price and seller are visible. |
be-trustworthyhumane-error-handling | issues | high | Submitting the empty Romanian login form focuses the field but yields browser text “Please fill in this field”, no aria-invalid, and no role=alert/error message. F28 medium: Login validation relies on an unlabeled native error in the wrong language. |
be-trustworthytrustworthy-input-assistance | issues | high | The login email field has an empty autocomplete value; homepage name/email newsletter inputs also have no autocomplete tokens. F29 medium: Sign-in and newsletter fields omit useful autocomplete tokens. |
be-trustworthysafe-commercial-and-account-flows | pass | high | The product page shows full price, instalment amount, seller, stock and return-related copy before the add-to-cart action; login offers federated alternatives without forced purchase continuity. |
be-sustainableoptimised-assets | issues | high | The image audit reports 19 oversized assets, 28 without srcset, 46 legacy-format images, and only 7 modern-format images. F30 medium: Image delivery wastes bytes and misses responsive formats. |
be-sustainableno-wasteful-work | issues | high | HAR transfers 1.85MB across 82 requests and Lighthouse estimates 288KB unused JavaScript while tracker/Sentry requests start during initial load. F31 medium: Initial loading includes avoidable code and background services. |
be-sustainablethird-party-and-media-budget | issues | high | Nine third-party origins are contacted and images account for 888,969 transferred bytes before the user interacts. F32 medium: Third-party and image costs are high for the first shopping view. |
be-agent-readystructured-agent-capabilities | not-applicable | high | No declared agent-facing scope or WebMCP capability was present; this emerging check is contextual and absence is not penalised. |
be-agent-readyon-device-inference | not-applicable | high | No summarisation, generation, or other inference task is part of the representative shopping flow; on-device AI is not applicable. |
be-memory-efficientno-leak-under-repeated-interaction | pass | high | After 10 repeated full-page scroll cycles, heap node count fell from 356,509 to 355,621 and self size fell from 21,686,575 to 21,588,207 bytes. |
be-memory-efficientbounded-footprint | pass | high | The measured homepage heap is about 21.7MB self size for a media-rich commerce page; constructor totals are stable and proportionate in the comparison. |
be-memory-efficientno-detached-dom-or-unbounded-listeners | pass | high | The before/after summaries show no growing Detached* constructor population and closures remain effectively flat (11,941 to 11,938) after repeated scrolling. |
Provenance
Canonical report: results/atomic/reports/0628-www_emag_ro.json
Report SHA-256: a4b8ca51eeac1260db926ad62884349e1643310476b6aaf62b590c493afa73b2
Local retained report: runs/2026-07-17T17-27-24-856Z/atomic-reports/www_emag_ro/2026-07-22T00-47-40-146Z/report.json
Local evidence root: runs/2026-07-17T17-27-24-856Z/atomic-reports/www_emag_ro/2026-07-22T00-47-40-146Z
Catalog SHA-256: sha256:78ccfdb2d483f4c57d9dafed80fd86c6265585a56457c8dcfddc254b80fb44d7
Raw screenshots, HARs, traces, heaps, and other browser artifacts are retained at the local evidence root and intentionally are not committed. Artifact paths in the canonical report are relative to that root.