Manifest position 297 · CrUX rank bucket 1000
https://accounts.google.com
Partial after retries
Atomic coverage is structurally complete, but 2 checks were blocked by credential requirements and 4 active tests were not run or could not be triggered reliably.
All 58 atomic check outcomes
| Principle / check | Status | Confidence | Evidence or reason |
|---|---|---|---|
respect-user-preferencesrespects-color-scheme | issues | high | The dark-emulated screenshot is pixel-equivalent in appearance to the light capture, while the DOM reports color-scheme: normal and a white body background. The separate signup route does render dark, showing inconsistent preference support across the account journey. F01 medium: The sign-in surface ignores the user’s dark color-scheme preference. |
respect-user-preferencesrespects-reduced-motion | pass | high | The reduced-motion probe confirmed the emulated preference and found zero active animations. |
respect-user-preferencesrespects-contrast | pass | high | Lighthouse contrast passed and the prefers-contrast screenshot retained visible text, input boundaries, and controls. |
implement-natural-interactionsview-transitions | not-run | high | A successful identifier-to-authentication state transition was not executed because it requires valid credentials; no equivalent safe transition was available in the Lite response. |
implement-natural-interactionsscroll-driven-animations | not-applicable | high | The audited identifier and signup forms contain no scroll-linked storytelling, parallax, carousel, or reveal motion. |
implement-natural-interactionsphysical-gestures | not-applicable | high | The audited forms expose no gesture-driven control or scroll-snap interaction. |
provide-guided-navigationscroll-state-aware-chrome | not-applicable | high | The short, single-task sign-in form has no persistent page chrome whose behavior needs to react to scroll position. |
provide-guided-navigationanchored-positioning | not-applicable | high | No tooltip, popover, context menu, or anchored overlay is present on the audited states. |
provide-guided-navigationdirects-attention | pass | high | Desktop/mobile screenshots show one H1, one outlined input, and one visually dominant Next action in reading order. |
maximize-content-reduce-noiseno-intrusive-interruptions | pass | high | Initial screenshots show the sign-in form immediately with no interstitial, consent wall, popup, or obscuring banner. |
maximize-content-reduce-noisesemantic-dismissible-primitives | not-applicable | high | No transient overlay, dialog, disclosure, or rich custom picker is present on the audited states. |
maximize-content-reduce-noisereduced-chrome | pass | high | The first viewport is dominated by the task form and explanatory copy; only compact footer links sit outside the form. |
adapt-to-the-form-factorresponsive-no-horizontal-scroll | pass | high | At 360x800, layout metrics report scrollWidth=clientWidth=360 and 0 horizontal overflow; viewport metadata is present. |
adapt-to-the-form-factorcomponent-level-responsiveness | pass | high | The form changes from a bordered centered desktop card to a full-width mobile composition without overflow or clipped controls. |
adapt-to-the-form-factorinput-modality-aware | issues | high | The active focus probe found :focus-visible=true on the email input, links, and Next button, but every control computed outline-style:none and box-shadow:none. The focused-link screenshot shows no visible change, and the Lite sign-in DOM has no main landmark. F02 high: Keyboard focus is not visibly distinguished on the sign-in controls. |
support-core-task-successclear-purpose-and-primary-action | pass | high | Both sign-in and signup states state their purpose in an H1 and present one visually dominant Next action. |
support-core-task-successprimary-flow-completion | blocked | high | A credentialed end-to-end sign-in requires a real Google account and must not be attempted with fabricated credentials. |
support-core-task-successclear-system-state-and-recovery | not-run | high | The synthetic untrusted click did not trigger the production validation flow, so error/success/retry states were not conclusively exercised. |
be-fast-and-stablegood-core-web-vitals | issues | high | Lighthouse measured LCP at 4.3 s and TTI at 6.1 s (performance 0.81). A separate unthrottled trace of the Lite response measured LCP at 0.843 s, so the regression is condition/variant-sensitive rather than universal. F03 medium: The full sign-in implementation misses the good LCP threshold under Lighthouse throttling. |
be-fast-and-stablevisual-stability | pass | high | Layout observer and Lighthouse both measured CLS 0 with no recorded shifts. |
be-fast-and-stableefficient-main-thread | pass | high | The raw trace recorded no >50 ms tasks and 0 ms TBT; Lighthouse TBT remained 193 ms. |
be-fast-and-stableefficient-resource-delivery | issues | high | The HAR shows three document redirects before the identifier page. Lighthouse reports a 686 ms longest network chain, including a 55 KB font, alongside the 4.3 s LCP. F04 low: Redirect and font/script dependency chains delay the full sign-in render. |
be-fast-and-stabletrim-unused-and-duplicate-code | issues | high | Lighthouse estimates 361 KiB of removable JavaScript and 165 KiB of removable CSS; two JavaScript bundles are 64–73% unused in the measured view. F05 medium: The identifier step ships substantial code not used during initial render. |
be-inclusivenames-roles-labels | pass | high | Lighthouse accessibility scored 1.0; label and button-name audits pass, and the Google image has alt text. |
be-inclusivesufficient-contrast | pass | high | Lighthouse color-contrast audit passes and normal/high-contrast screenshots remain legible. |
be-inclusivestructure-and-focus | issues | high | The active focus probe found :focus-visible=true on the email input, links, and Next button, but every control computed outline-style:none and box-shadow:none. The focused-link screenshot shows no visible change, and the Lite sign-in DOM has no main landmark. F02 high: Keyboard focus is not visibly distinguished on the sign-in controls. |
be-inclusivelegible-text | pass | high | Desktop and 320/360 px screenshots show unclipped text, readable line lengths, and clear hierarchy. |
be-inclusivezoom-reflow-targets-and-media | pass | high | Viewport scaling is not disabled, mobile reflow has no horizontal overflow, and the page contains no audio/video requiring alternatives. |
follow-best-practicesno-console-errors | pass | high | Lighthouse reports no browser console errors. |
follow-best-practicessound-document-and-assets | pass | high | DOM/Lighthouse confirm HTML doctype, UTF-8 charset, viewport metadata, and correctly dimensioned image with matching 2x display size. |
follow-best-practicesbrowser-platform-hygiene | pass | high | Lighthouse best-practices scored 1.0; no permission prompts appeared and the bfcache exclusions are security-related no-store/WebAuthn constraints. |
be-discoverabletitle-and-description | issues | high | The DOM probe and Lighthouse both found a descriptive title but no meta[name=description]. F08 low: The public sign-in entry has no meta description. |
be-discoverablecrawlable-and-mobile-friendly | pass | high | Lighthouse confirms crawlable anchors, a valid robots.txt, indexing allowed, and viewport metadata; link labels are descriptive. |
be-discoverablecanonical-and-indexing-signals | pass | high | The entry and final sign-in documents returned successful/expected redirect statuses, robots.txt allows the page, and no accidental noindex policy was found. A canonical is not appropriate for session-specific authentication continuation URLs. |
be-discoverablestructured-and-shareable-metadata | not-applicable | high | The authentication form is not a public article, product, event, organization profile, or other shareable rich entity. |
be-private-and-securesecure-transport-and-headers | pass | high | HTTPS, HSTS, nosniff, X-Frame-Options:DENY, CSP nonces/Trusted Types, and a Secure HttpOnly __Host- cookie were directly observed. |
be-private-and-securedata-minimisation-and-third-parties | pass | high | Tracker probe found no known tracker domains; HAR recorded only two low-byte third-party requests and secrets scan found no exposed credentials. |
be-private-and-securein-context-permissions-and-modern-auth | pass | high | No on-load permission activity was observed; Lighthouse identifies WebAuthentication API use, corroborating phishing-resistant authentication support. |
be-private-and-securedefensive-browser-policies | issues | high | The headers probe found no Referrer-Policy header and a CSP policy containing unsafe-eval. Strong compensating controls are present, including HSTS, X-Frame-Options:DENY, nosniff, nonces, and require-trusted-types-for script. F06 medium: The authentication response has gaps in explicit browser-enforced policy. |
be-resilientprogressive-enhancement | pass | high | Discoverability raw fetch is HTTP 200, not a JS shell, retains title/H1, and exposes 76% of rendered content without JavaScript. |
be-resilientresilient-runtime-behaviour | pass | high | No overlays can be clipped, no console/runtime errors were observed, and the simple server-rendered state remained stable. |
be-resilientoffline-and-installable | not-applicable | high | Authentication is intrinsically online and should not be installable or claim offline completion. |
be-resilientnetwork-and-http-failure-states | not-run | high | Offline and HTTP-failure injection was not run against this live authentication endpoint to avoid disrupting or misrepresenting security-sensitive behavior. |
be-internationalisedlang-dir-and-logical-properties | pass | high | The DOM declares lang=en-US and dir=ltr, and both sign-in and signup expose a language selector. |
be-internationalisedlocale-aware-data | not-applicable | high | The audited states display no dates, numbers, currency, duration, or calendar data to localize. |
be-internationalisedtime-zone-correctness | not-applicable | high | The audited states expose no time or recurring-event concepts. |
be-trustworthyno-dark-patterns | pass | high | Initial sign-in/signup screenshots show no preselected consent, upsell, continuity trap, disguised ad, or confirmshaming; recovery and account creation are plainly labelled. |
be-trustworthyhumane-error-handling | not-run | high | The production validation state could not be triggered reliably with the generic untrusted interaction probe, so timing and accessible announcement were not judged. |
be-trustworthytrustworthy-input-assistance | pass | high | The identifier field exposes autocomplete=username and visible Email or phone labelling. |
be-trustworthysafe-commercial-and-account-flows | blocked | high | Credentialed sign-in, reauthentication, account management, and cancellation/reversal states require an authorized test account. |
be-sustainableoptimised-assets | pass | high | The only image is a dimensioned 3.7 KB 2x logo, displayed at exactly half its natural dimensions; no oversized assets were found. |
be-sustainableno-wasteful-work | issues | high | Lighthouse estimates 361 KiB unused JavaScript and 165 KiB unused CSS on a screen whose core UI is one text field and one primary button. F07 low: Unused client code adds avoidable transfer, parse, and memory cost to a simple identifier form. |
be-sustainablethird-party-and-media-budget | pass | high | The Lite HAR transfers 156.6 KB with 5.2 KB third-party bytes and no audio/video/autoplay media. |
be-agent-readystructured-agent-capabilities | not-applicable | high | Exposing sign-in or account-creation actions as general agent tools would be inappropriate for this sensitive authentication surface; no declared agent-facing intent exists. |
be-agent-readyon-device-inference | not-applicable | high | The deterministic sign-in form has no summarization or language-model task that would benefit from on-device inference. |
be-memory-efficientno-leak-under-repeated-interaction | pass | medium | After ten focus/input/clear/blur cycles, heap self-size changed from 7,936,758 to 7,941,728 bytes (+4,970 bytes, 0.06%), with no unbounded-growth signal. |
be-memory-efficientbounded-footprint | pass | medium | The Lite page heap is about 7.94 MB with 178 live DOM elements and 4.77 MB JS heap used, proportionate to the form. |
be-memory-efficientno-detached-dom-or-unbounded-listeners | pass | medium | The before/after summaries show only +13 native nodes after ten cycles and no Detached* constructor among the reported top constructors. |
Provenance
Canonical report: results/atomic/reports/0297-accounts_google_com.json
Report SHA-256: 58bf157b71845783d2a66083b504a3ea48a46b5078f0c37188c8d7af9c448936
Local retained report: runs/2026-07-17T17-27-24-856Z/atomic-reports/accounts_google_com/2026-07-27T19-14-23-318Z/report.json
Local evidence root: runs/2026-07-17T17-27-24-856Z/atomic-reports/accounts_google_com/2026-07-27T19-14-23-318Z
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.