Manifest position 267 · CrUX rank bucket 1000
https://www.amazon.co.uk
Coverage complete
Atomic coverage complete across 58 checks and six representative public paths. Purchase was not irreversibly submitted; checkout entry and sign-in were inspected.
All 58 atomic check outcomes
| Principle / check | Status | Confidence | Evidence or reason |
|---|---|---|---|
respect-user-preferencesrespects-color-scheme | issues | high | The dark-emulated screenshot remains the same white/light catalogue as the normal capture; DOM computed styles report color-scheme: normal and product CSS contains no prefers-color-scheme rule. F01 medium: The experience ignores the user’s dark colour preference. |
respect-user-preferencesrespects-reduced-motion | issues | high | The reduced-motion probe reports 34 running _aXZlL_loadingShimmer animations, each 1500 ms, while matchMedia confirms prefers-reduced-motion: reduce; product CSS has no matching media rule. F02 medium: Loading shimmer motion continues under reduced-motion. |
respect-user-preferencesrespects-contrast | issues | high | Product CSS contains no prefers-contrast or forced-colors rules; controls rely heavily on custom sprite colours. Lighthouse found no ordinary text contrast failure, so this is a preference-support gap rather than a general contrast failure. F03 low: No explicit high-contrast or forced-colours adaptation is present. |
implement-natural-interactionsview-transitions | issues | high | DOM/CSS inspection found no view-transition CSS or scroll animation timeline; the add-to-basket screencast produced only two abrupt state frames. F04 low: State changes do not use modern declarative transition primitives. |
implement-natural-interactionsscroll-driven-animations | issues | high | DOM/CSS inspection found no view-transition CSS or scroll animation timeline; the add-to-basket screencast produced only two abrupt state frames. F04 low: State changes do not use modern declarative transition primitives. |
implement-natural-interactionsphysical-gestures | pass | high | Product CSS contains 10 scroll-snap/overscroll-behavior signals and no custom pointer-move gesture was observed. |
provide-guided-navigationscroll-state-aware-chrome | issues | high | Homepage/product CSS has no @container scroll-state, anchor-name, position-anchor, or position-try signals despite numerous role=dialog/menu surfaces. F05 low: Global navigation does not use scroll-state or anchor positioning for contextual chrome. |
provide-guided-navigationanchored-positioning | issues | high | Homepage/product CSS has no @container scroll-state, anchor-name, position-anchor, or position-try signals despite numerous role=dialog/menu surfaces. F05 low: Global navigation does not use scroll-state or anchor positioning for contextual chrome. |
provide-guided-navigationdirects-attention | pass | high | Search results, product title/price, skip navigation and Add to basket controls provide clear visual and keyboard attention cues. |
maximize-content-reduce-noiseno-intrusive-interruptions | issues | high | Fresh-profile homepage and product screenshots show a fixed consent form covering roughly the lower 40% of a 1000 px viewport. The overlay is dismissible and offers equal Accept/Decline controls, but content is substantially obscured. F06 medium: The first-visit consent panel obscures a large share of the core page. |
maximize-content-reduce-noisesemantic-dismissible-primitives | issues | high | The platform probe found 25 role-dialog/aria-modal surfaces but zero dialog and zero popover elements; the consent UI is a fixed form. F07 medium: Overlay infrastructure is largely custom rather than native semantic primitives. |
maximize-content-reduce-noisereduced-chrome | issues | high | Homepage screenshots devote two navigation bars plus multiple full-height promotional rails to chrome; at 360 px the desktop canvas is clipped and adjacent cards remain visible off-screen. F08 high: Dense global chrome and horizontal card rails crowd primary content. |
adapt-to-the-form-factorresponsive-no-horizontal-scroll | issues | high | All home/search/product/cart DOM captures report hasViewportMeta=false. At 360x800 the homepage layout reports a 1000 px inner canvas and 640 px horizontal overflow; the screenshot shows clipped desktop navigation and cards. F09 critical: Public templates omit the viewport meta tag and render a desktop canvas on phones. |
adapt-to-the-form-factorcomponent-level-responsiveness | issues | high | DOM/CSS probes found no @container rules, while the same catalogue and product modules remain desktop-sized in narrow captures. F10 medium: Reusable modules do not demonstrate component-level adaptation. |
adapt-to-the-form-factorinput-modality-aware | issues | high | Lighthouse target-size failed on two groups; the product focus probe counted 19 sub-24 px targets and showed several global links/buttons with outline:none after programmatic focus, including Add to basket shortcuts. F11 high: Some controls are undersized or lose visible keyboard focus. |
support-core-task-successclear-purpose-and-primary-action | pass | high | Search clearly labels Results; product exposes title, price, stock and Add to basket; cart clearly explains its empty state. |
support-core-task-successprimary-flow-completion | pass | high | Search, product, basket and sign-in entry were reached in sequence; visible controls and stable URLs support discovery through checkout entry without an avoidable dead end before credential entry. |
support-core-task-successclear-system-state-and-recovery | pass | high | Empty basket explains state and offers shopping/sign-in choices; invalid product screenshot gives an explicit apology and homepage recovery link. |
be-fast-and-stablegood-core-web-vitals | issues | high | Lighthouse performance scored 0.37 with FCP 6.5 s, LCP 15.2 s, TBT 840 ms and TTI 19.1 s. The raw-CDP trace corroborated FCP 4.51 s, LCP 4.57 s and two long tasks. F12 high: Cold-load performance is poor. |
be-fast-and-stablevisual-stability | pass | high | Layout observer reported CLS 0 with no recorded shifts; Lighthouse also reports CLS 0. |
be-fast-and-stableefficient-main-thread | issues | high | Lighthouse performance scored 0.37 with FCP 6.5 s, LCP 15.2 s, TBT 840 ms and TTI 19.1 s. The raw-CDP trace corroborated FCP 4.51 s, LCP 4.57 s and two long tasks. F12 high: Cold-load performance is poor. |
be-fast-and-stableefficient-resource-delivery | issues | high | HAR recorded 127 requests and 2.89 MB transferred, including 1.06 MB script, 735 KB fonts, 10 grounded render-blocking candidates, and 2.43 MB from non-main origins; a WAF challenge script alone transferred 591 KB. F13 high: The load ships a heavy blocking resource graph. |
be-fast-and-stabletrim-unused-and-duplicate-code | issues | high | HAR recorded 127 requests and 2.89 MB transferred, including 1.06 MB script, 735 KB fonts, 10 grounded render-blocking candidates, and 2.43 MB from non-main origins; a WAF challenge script alone transferred 591 KB. F13 high: The load ships a heavy blocking resource graph. |
be-inclusivenames-roles-labels | issues | high | Lighthouse found aria-allowed-attr, aria-prohibited-attr and label-content-name-mismatch failures; image probes found 2 missing alts on home, 7 on search and 38 on product. F14 medium: Accessible names and ARIA semantics contain defects. |
be-inclusivesufficient-contrast | pass | high | Lighthouse accessibility scored 0.90 and did not report a colour-contrast failure; screenshots show readable primary text/control contrast. |
be-inclusivestructure-and-focus | issues | high | Lighthouse reports a malformed agent accessibility tree; focus probing showed multiple visible controls with no outline, and the consent Accept/Decline inputs computed outline:none. F15 high: Keyboard and accessibility-tree quality is inconsistent. |
be-inclusivelegible-text | pass | high | Desktop search/product screenshots show readable line lengths, hierarchy and unclipped product copy. |
be-inclusivezoom-reflow-targets-and-media | issues | high | Every representative DOM lacked viewport meta, narrow layouts retained 980–1000 px canvases, and Lighthouse failed target size. F16 high: Zoom/reflow support is undermined by the missing viewport and small targets. |
follow-best-practicesno-console-errors | issues | high | Lighthouse errors-in-console failed with one recorded error. F17 medium: The homepage logs browser errors. |
follow-best-practicessound-document-and-assets | issues | high | Lighthouse found six incorrect-aspect-ratio instances; image probes found 41/42 home images and 89/100 product images without both dimensions. Doctype and UTF-8 charset pass. F18 medium: Image and document asset hygiene is inconsistent. |
follow-best-practicesbrowser-platform-hygiene | issues | high | Lighthouse deprecations failed with two items; browser console errors also occurred. F19 medium: Deprecated APIs and inspector hygiene failures remain. |
be-discoverabletitle-and-description | issues | high | DOM metadata summary shows search and cart have no meta description or canonical; search also has no structured/share metadata. Lighthouse reports one non-crawlable anchor. F20 high: Search and basket templates omit essential metadata and indexing signals. |
be-discoverablecrawlable-and-mobile-friendly | issues | high | DOM metadata summary shows search and cart have no meta description or canonical; search also has no structured/share metadata. Lighthouse reports one non-crawlable anchor. F20 high: Search and basket templates omit essential metadata and indexing signals. |
be-discoverablecanonical-and-indexing-signals | issues | high | DOM metadata summary shows search and cart have no meta description or canonical; search also has no structured/share metadata. Lighthouse reports one non-crawlable anchor. F20 high: Search and basket templates omit essential metadata and indexing signals. |
be-discoverablestructured-and-shareable-metadata | issues | high | Product probe found zero JSON-LD and zero Open Graph tags despite a visible named product, price, rating and availability. F21 high: Product entities lack structured and social-preview metadata. |
be-private-and-securesecure-transport-and-headers | issues | high | HTTPS passes, but the headers primitive found no CSP, HSTS, nosniff, frame protection, Referrer-Policy or Permissions-Policy. Cookie audit found 7/8 cookies insecure and auth-like session-id/session-id-time not HttpOnly. F22 high: Core browser security headers and cookie protections are missing. |
be-private-and-securedata-minimisation-and-third-parties | issues | high | Consent copy names 129 third parties. HAR attributes 112/127 requests and 2.43 MB to non-main origins, though the tracker primitive found no known third-party tracker domains and secrets scan found no credentials. F23 medium: The first load has a broad third-party/data-processing footprint. |
be-private-and-securein-context-permissions-and-modern-auth | pass | high | No notification/geolocation prompt appeared on load; sign-in markup includes WebAuthn/passkey support strings and username autocomplete. |
be-private-and-securedefensive-browser-policies | issues | high | HTTPS passes, but the headers primitive found no CSP, HSTS, nosniff, frame protection, Referrer-Policy or Permissions-Policy. Cookie audit found 7/8 cookies insecure and auth-like session-id/session-id-time not HttpOnly. F22 high: Core browser security headers and cookie protections are missing. |
be-resilientprogressive-enhancement | issues | high | Discoverability captures report homepage 1% raw/rendered word coverage (HTTP 202) and search 0%, both isJsShell=true; by contrast product content is 96% present, showing the resilient product pattern is achievable. F24 high: Homepage and search content collapse for non-JavaScript crawlers. |
be-resilientresilient-runtime-behaviour | pass | high | Consent, search filters, product commerce panel and cart state remained visible and were not cut off in desktop captures. |
be-resilientoffline-and-installable | issues | high | Platform probe found an active service-worker controller but no manifest link; no installable identity or offline fallback was exposed on the commerce entry page. F25 medium: No installable/offline product surface was exposed. |
be-resilientnetwork-and-http-failure-states | pass | high | Invalid product path presents a clear, compact error and a working route back to the homepage. |
be-internationalisedlang-dir-and-logical-properties | pass | high | Rendered documents use lang=en-gb and platform probe found logical CSS properties; UK currency/date strings are localized. |
be-internationalisedlocale-aware-data | pass | high | Product/search pages consistently render UK pound currency and localized delivery-date phrasing. |
be-internationalisedtime-zone-correctness | not-applicable | high | The selected public shopping archetypes expose relative delivery dates but no cross-time-zone event or recurring-time concept to evaluate. |
be-trustworthyno-dark-patterns | pass | high | Consent gives Accept, Decline and Customise together without confirmshaming; product price, APR disclosure, stock, returns and seller are visible. |
be-trustworthyhumane-error-handling | pass | high | Invalid URL error is plain-language and actionable; sign-in labels the required identifier before continuation. |
be-trustworthytrustworthy-input-assistance | pass | high | Sign-in identifier uses autocomplete=username; search has an accessible label and intentionally disables unrelated autofill. |
be-trustworthysafe-commercial-and-account-flows | pass | high | Product shows total price, stock, delivery, returns, seller and finance disclosure before purchase; account creation and help are visible at sign-in. |
be-sustainableoptimised-assets | issues | high | Home uses 35 legacy images and 40 images without srcset; product uses 91 legacy images, 51 oversized images and 89 missing dimensions. F26 high: Image delivery is wasteful across key templates. |
be-sustainableno-wasteful-work | issues | high | HAR shows 127 requests/2.89 MB, 24 scripts/1.06 MB and 112 non-main-origin requests; Lighthouse TBT is 840 ms and TTI 19.1 s. F27 high: Background and third-party work exceeds a proportionate first-load budget. |
be-sustainablethird-party-and-media-budget | issues | high | HAR shows 127 requests/2.89 MB, 24 scripts/1.06 MB and 112 non-main-origin requests; Lighthouse TBT is 840 ms and TTI 19.1 s. F27 high: Background and third-party work exceeds a proportionate first-load budget. |
be-agent-readystructured-agent-capabilities | issues | high | The public product/search flows are relevant to agents, but the platform probe found no WebMCP surface and Lighthouse found the agent accessibility tree malformed. F28 low: Agentic commerce capabilities are not exposed as structured tools. |
be-agent-readyon-device-inference | issues | high | Chrome exposes LanguageModel and Summarizer globals in the probe, but the page registered no observable on-device summarisation or product-comparison capability. F29 low: No page-owned on-device inference enhancement was observed. |
be-memory-efficientno-leak-under-repeated-interaction | pass | high | Same-page 10-cycle scroll probe kept DOM nodes exactly 1612 and animations exactly 3; JS heap rose only 4.63%, not evidence of unbounded retained growth. |
be-memory-efficientbounded-footprint | pass | medium | Heap baseline was 393,510 nodes and 23.9 MB self size; proportionate for this complex catalogue page, though optimisation remains desirable. |
be-memory-efficientno-detached-dom-or-unbounded-listeners | pass | medium | Across ten same-page scroll cycles DOM nodes and animation count were flat; separate snapshots varied only 1.01% in node count, with no accumulating visible session state. |
Provenance
Canonical report: results/atomic/reports/0267-www_amazon_co_uk.json
Report SHA-256: 15b83086540edfe69c62158eabc31cbdd5daeca87743578fff7045659a880dcb
Local retained report: runs/2026-07-17T17-27-24-856Z/atomic-reports/www_amazon_co_uk/2026-07-19T14-04-54-798Z/report.json
Local evidence root: runs/2026-07-17T17-27-24-856Z/atomic-reports/www_amazon_co_uk/2026-07-19T14-04-54-798Z
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.