Manifest position 479 · CrUX rank bucket 1000
https://www.bing.com
Coverage complete
Coverage complete. Representative web, image, news, mobile, preference, security, performance, no-JS, error and memory conditions were judged.
All 58 atomic check outcomes
| Principle / check | Status | Confidence | Evidence or reason |
|---|---|---|---|
respect-user-preferencesrespects-color-scheme | issues | high | The dark-preference capture keeps the same white search and consent surfaces as the default capture; computed color-scheme is normal and the CSS probe found no prefers-color-scheme theming rule. F-01 medium: The homepage does not expose a user-preference-driven dark theme. |
respect-user-preferencesrespects-reduced-motion | issues | high | Under prefers-reduced-motion: reduce, getAnimations() still reported a running slide-up animation with a 2000 ms duration, and no reduced-motion rule was found. F-02 medium: Motion continues when reduced motion is requested. |
respect-user-preferencesrespects-contrast | pass | medium | respects-contrast: Desktop, dark-preference and forced-colors captures plus the media/CSS probe directly exercised platform preference conditions. |
implement-natural-interactionsview-transitions | issues | high | CSS inspection found no view-transition rules across the homepage while navigation changes between search verticals are meaningful state/route swaps. F-03 low: Route and state changes do not use declarative View Transitions. |
implement-natural-interactionsscroll-driven-animations | issues | high | CSS inspection found no animation-timeline rules on the horizontally scrolling news/image surfaces. F-04 low: Scrollable feeds and carousels do not expose declarative scroll-driven treatment. |
implement-natural-interactionsphysical-gestures | issues | high | Homepage and news screenshots show horizontal card rails with custom arrows; CSS inspection found no scroll-snap primitive. F-05 low: Horizontal carousels rely on custom arrow controls without native scroll-snap evidence. |
provide-guided-navigationscroll-state-aware-chrome | issues | high | The homepage has a long Discover surface and news feed, but inspection found no scroll-state query or progress treatment. F-06 low: Long content surfaces provide static rather than scroll-state-aware chrome. |
provide-guided-navigationanchored-positioning | issues | high | The DOM exposes custom tooltip/account/visual-search overlays, while CSS inspection found no anchor-name or position-try usage. F-07 low: Transient menus and tooltips use custom positioning rather than CSS anchors. |
provide-guided-navigationdirects-attention | issues | high | Search, Images and News screenshots show immediate template swaps with only tab underline changes and no transition/highlight cue. F-08 low: Navigation state changes give limited directional or destination feedback. |
maximize-content-reduce-noiseno-intrusive-interruptions | issues | high | The rendered 780 px viewport shows the consent dialog consuming roughly the lower half of the page; desktop captures also cover the lower content rail. F-09 medium: The first-load consent panel obscures a large portion of content. |
maximize-content-reduce-noisesemantic-dismissible-primitives | issues | high | The DOM probe found the cookie, image search and trivia overlays as div[role=dialog], and Lighthouse reported an unnamed dialog. F-10 low: Several overlays are ad-hoc role=dialog containers rather than native dialog/popover primitives. |
maximize-content-reduce-noisereduced-chrome | pass | medium | reduced-chrome: First-load desktop/mobile screenshots and overlay DOM show the content/chrome balance and consent treatment. |
adapt-to-the-form-factorresponsive-no-horizontal-scroll | issues | high | The mobile layout primitive measured scrollWidth 769 px against visualViewportWidth 360 px, or 409 px horizontal overflow; the screenshot clips the search box and rails. F-11 high: The homepage overflows severely at a 360 px viewport. |
adapt-to-the-form-factorcomponent-level-responsiveness | issues | high | CSS inspection reported container-type: normal and no @container rules; the homepage search box and card rail remain desktop-width and clip on mobile. F-12 low: Reusable search and card components show no container-query adaptation. |
adapt-to-the-form-factorinput-modality-aware | issues | high | The focus/target probe measured the visual-search and clear controls at 24x24 px and the menu at 20 px wide; focused search textarea and image-search button computed with no outline or box shadow. F-13 high: Important controls have undersized targets or no visible keyboard focus. |
support-core-task-successclear-purpose-and-primary-action | pass | medium | clear-purpose-and-primary-action: The entry page, direct query results, image results, news results and an invalid URL were walked as representative task and recovery states. |
support-core-task-successprimary-flow-completion | pass | medium | primary-flow-completion: The entry page, direct query results, image results, news results and an invalid URL were walked as representative task and recovery states. |
support-core-task-successclear-system-state-and-recovery | issues | high | Navigating to /this-path-should-not-exist-web-uplift rendered the normal Bing homepage rather than a clear 404 explanation or recovery choice. F-14 medium: Unknown URLs do not present a clear not-found recovery state. |
be-fast-and-stablegood-core-web-vitals | issues | high | Lighthouse measured LCP at 4.646 s (score 0.34), although the separate trace measured 1.936 s; the slower throttled result establishes a common constrained-device risk. F-15 high: Laboratory LCP is outside the good range. |
be-fast-and-stablevisual-stability | pass | medium | visual-stability: Trace, Lighthouse, layout metrics, HAR and image audit directly measured loading, stability, main-thread and delivery behavior. |
be-fast-and-stableefficient-main-thread | pass | medium | efficient-main-thread: Trace, Lighthouse, layout metrics, HAR and image audit directly measured loading, stability, main-thread and delivery behavior. |
be-fast-and-stableefficient-resource-delivery | issues | high | HAR recorded 180 requests, including 18 stylesheets and ten parser-inserted VeryHigh-priority stylesheet candidates; Lighthouse failed LCP discovery and network dependency insights. F-16 medium: The critical network path is fragmented and request-heavy. |
be-fast-and-stabletrim-unused-and-duplicate-code | issues | high | HAR recorded 75 scripts plus 42 XHR/fetch requests and 129 third-party-origin requests; Lighthouse also reported missing source maps for large first-party JavaScript. F-17 medium: The homepage ships a large, fragmented script surface for its core search task. |
be-inclusivenames-roles-labels | issues | high | Lighthouse failed aria-dialog-name; the image audit found 6 images without alt text, and the probe found an empty H1 and unnamed custom controls. F-18 high: Some dialogs and images lack accessible names or alternatives. |
be-inclusivesufficient-contrast | pass | medium | sufficient-contrast: Lighthouse accessibility audits, forced-colors/mobile captures and a keyboard focus/semantic probe directly tested inclusion. |
be-inclusivestructure-and-focus | issues | high | The DOM probe found an empty H1 and focused primary search/image controls with no outline or box shadow; Lighthouse reported a malformed accessibility tree. F-19 high: Heading semantics and focus visibility are inconsistent. |
be-inclusivelegible-text | pass | medium | legible-text: Lighthouse accessibility audits, forced-colors/mobile captures and a keyboard focus/semantic probe directly tested inclusion. |
be-inclusivezoom-reflow-targets-and-media | issues | high | The 360 px page has 409 px horizontal overflow, several targets are 20-24 px, and Lighthouse failed meta-viewport because scaling is restricted below its threshold. F-20 high: Reflow, zoom and target-size behavior fail the inclusive mobile baseline. |
follow-best-practicesno-console-errors | issues | high | Lighthouse failed the errors-in-console audit, despite the later in-page error array being empty, indicating load-time errors not retained by the page probe. F-21 medium: The page logs browser errors during Lighthouse loading. |
follow-best-practicessound-document-and-assets | issues | high | The image audit found 25 of 32 images without width/height attributes, creating latent layout-shift risk even though the sampled CLS was low. F-22 medium: Many content images omit intrinsic dimensions. |
follow-best-practicesbrowser-platform-hygiene | issues | high | Lighthouse reported one deprecated API warning, missing source maps for large first-party JavaScript and DevTools inspector issues. F-23 medium: Deprecated API, source-map and inspector issues remain. |
be-discoverabletitle-and-description | pass | medium | title-and-description: Raw-vs-rendered discoverability, robots, DOM metadata and Lighthouse SEO evidence directly inspected crawl/index signals. |
be-discoverablecrawlable-and-mobile-friendly | issues | high | The probe found 6 javascript: links and Lighthouse failed crawlable-anchors and meta-viewport; robots.txt disallows /search. F-24 high: Search links and mobile crawler signals are weakened. |
be-discoverablecanonical-and-indexing-signals | pass | medium | canonical-and-indexing-signals: Raw-vs-rendered discoverability, robots, DOM metadata and Lighthouse SEO evidence directly inspected crawl/index signals. |
be-discoverablestructured-and-shareable-metadata | issues | high | The DOM probe found zero JSON-LD blocks while the page represents the Bing organization/search service; Open Graph metadata is present. F-25 low: The homepage has social tags but no structured entity data. |
be-private-and-securesecure-transport-and-headers | issues | high | Headers include HTTPS, HSTS and CSP, but omit X-Content-Type-Options and Referrer-Policy; cookies reported MUIDB, _EDGE_V and _EDGE_S without Secure and pervasive SameSite=None. F-26 high: Cookie and header hardening is incomplete. |
be-private-and-securedata-minimisation-and-third-parties | issues | high | HAR recorded 129 third-party-origin requests transferring 617,687 bytes; trackers enumerated 8 third-party origins and the consent dialog states advertising/analytics use. F-27 medium: The default homepage has a broad third-party and analytics footprint. |
be-private-and-securein-context-permissions-and-modern-auth | not-applicable | high | Authentication is delegated to Microsoft identity outside the audited host, and no permission prompt fired on the audited Bing surfaces; the host-owned portion is therefore not applicable to these unauthenticated paths. |
be-private-and-securedefensive-browser-policies | issues | high | HSTS and SAMEORIGIN are present, but Referrer-Policy and nosniff are absent; Permissions-Policy only disables unload and CSP allows wasm-unsafe-eval. F-28 medium: Defensive browser policies are only partially configured. |
be-resilientprogressive-enhancement | issues | high | Discoverability measured only 14% rendered-word coverage in raw HTML; the crawler screenshot is a blank dark surface while the rendered screenshot contains search, navigation and content. F-29 high: Most rendered homepage content is unavailable without JavaScript. |
be-resilientresilient-runtime-behaviour | pass | medium | resilient-runtime-behaviour: No-JS crawler comparison, multiple routes and an invalid route directly exercised progressive content and recovery behavior. |
be-resilientoffline-and-installable | not-applicable | high | Bing search is intrinsically network-dependent; an installable offline experience is not required for the representative public search surfaces. |
be-resilientnetwork-and-http-failure-states | issues | high | The deliberately invalid URL rendered the homepage with no not-found state, so users and agents cannot distinguish a bad route from successful navigation. F-30 medium: The invalid-route experience hides the failure rather than explaining recovery. |
be-internationalisedlang-dir-and-logical-properties | pass | medium | lang-dir-and-logical-properties: DOM metadata and representative en-GB content were inspected for language, direction and locale-sensitive content. |
be-internationalisedlocale-aware-data | not-applicable | high | The audited homepage/search/image/news surfaces contain no user-facing dates, currencies, durations or entered locale-sensitive data requiring formatting. |
be-internationalisedtime-zone-correctness | not-applicable | high | The representative audited surfaces expose no scheduled events or time-zone-sensitive values. |
be-trustworthyno-dark-patterns | pass | medium | no-dark-patterns: Consent, search, image search and unauthenticated account entry surfaces were visually and semantically inspected. |
be-trustworthyhumane-error-handling | not-applicable | high | The representative core search input accepts arbitrary text and has no required/invalid state to trigger; no validation flow genuinely applies. |
be-trustworthytrustworthy-input-assistance | not-applicable | high | The core query is free-form search text where personal-data autofill tokens would be inappropriate; autocomplete=off does not obstruct an address, payment or identity task. |
be-trustworthysafe-commercial-and-account-flows | not-applicable | high | No checkout, subscription or account-management flow exists on the audited unauthenticated Bing surfaces; sign-in is delegated outside the host. |
be-sustainableoptimised-assets | issues | high | The image audit found 20 images without srcset and 25 without dimensions; HAR also transferred both a 234,651-byte WebP and a 186,024-byte AVIF homepage background candidate. F-31 medium: Responsive image delivery is incomplete despite modern formats. |
be-sustainableno-wasteful-work | pass | medium | no-wasteful-work: HAR and image evidence directly measured bytes, request mix, formats, responsive delivery and third-party weight. |
be-sustainablethird-party-and-media-budget | issues | high | 129 of 180 requests were classified as third-party-origin and transferred 617,687 bytes, nearly half the 1.316 MB total. F-32 medium: Third-party work is disproportionate to the core search entry task. |
be-agent-readystructured-agent-capabilities | issues | high | Lighthouse agentic-browsing scored 0.50 and failed accessibility-tree quality; no-JS coverage was 14%, leaving agents dependent on brittle rendered-page interaction. F-33 low: The search service offers weak machine-readable capability exposure. |
be-agent-readyon-device-inference | not-applicable | high | The audited public search pages provide a server-backed search service; no bounded on-device summarisation or language-model task was evident where local inference is clearly preferable. |
be-memory-efficientno-leak-under-repeated-interaction | issues | low | Baseline heap summary had 205,144 nodes and 12,961,305 self-size bytes; after ten focus/input/clear/blur cycles the post summary had 1,093,682 nodes and 62,029,465 bytes. Independent page loads reduce confidence, but the 4.8x retained-size signal is too large to dismiss. F-34 medium: Heap growth after repeated query edits warrants leak investigation. |
be-memory-efficientbounded-footprint | issues | medium | The post-interaction heap summary retained 1.09 million nodes, 4.72 million edges and 62.0 MB self size after a small repeated query interaction. F-35 medium: The post-interaction footprint is heavy for the homepage shell. |
be-memory-efficientno-detached-dom-or-unbounded-listeners | issues | low | No Detached* constructor appeared in the compact summaries, but closures rose from 9,187 to 65,148 across the baseline/post captures alongside major node growth. F-36 low: Listener/timer accumulation could not be ruled out and the object population grew sharply. |
Provenance
Canonical report: results/atomic/reports/0479-www_bing_com.json
Report SHA-256: 03fb354e3c2166a0ec9a1f1335c6884183d1561e09eddeb0acf706486550f255
Local retained report: runs/2026-07-17T17-27-24-856Z/atomic-reports/www_bing_com/2026-07-21T02-00-52-738Z/report.json
Local evidence root: runs/2026-07-17T17-27-24-856Z/atomic-reports/www_bing_com/2026-07-21T02-00-52-738Z
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.