Manifest position 849 · CrUX rank bucket 1000
https://m.tabor.ru
Coverage complete
Coverage-complete audit of public unauthenticated templates. Authenticated profile, messaging, matching, payment, and account-deletion flows were not accessible without credentials; these exclusions are explicit in paths/report.md.
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 pixel-identical in appearance to the light capture; body and form surfaces remain fixed pale cream/white and the DOM CSS contains no color-scheme or prefers-color-scheme rule. F-01 medium: The site ignores the user’s dark color-scheme preference. |
respect-user-preferencesrespects-reduced-motion | pass | high | Under prefers-reduced-motion: reduce, getAnimations() returned no active animations; the audited pages use no non-essential motion or auto-advance. |
respect-user-preferencesrespects-contrast | issues | high | The prefers-contrast: more capture does not adapt the palette, and axe measured five failures including 1.98:1 for the h1 and 2.49:1 for the cookie button. F-02 high: Text and controls do not adapt to increased contrast and several fail WCAG contrast. |
implement-natural-interactionsview-transitions | issues | medium | Navigation among home, registration, recovery, reviews, and help performs abrupt full-document swaps; DOM/CSS inspection found no ::view-transition rules or view-transition-name usage. F-03 low: Primary route changes are abrupt and provide no spatial continuity. |
implement-natural-interactionsscroll-driven-animations | not-applicable | high | No parallax, scrollytelling, entry reveal, carousel, or other scroll-linked motion exists on the audited public templates. |
implement-natural-interactionsphysical-gestures | not-applicable | high | The public unauthenticated templates expose ordinary links/forms only, with no swipe, pull-to-refresh, draggable, snap-carousel, or gesture-driven component. |
provide-guided-navigationscroll-state-aware-chrome | issues | medium | The long reviews listing has no sticky context, progress, or scroll-state response; its only menu is at the top while pagination appears after the list. F-04 low: Long content offers no scroll-state orientation or persistent navigation. |
provide-guided-navigationanchored-positioning | not-applicable | high | No tooltip, attached popover, dropdown, or edge-sensitive overlay is exposed on the audited unauthenticated templates. |
provide-guided-navigationdirects-attention | pass | medium | Headings, explicit action labels, numbered registration progress, back links, and review pagination identify the current context and next location without ambiguous icon-only navigation. |
maximize-content-reduce-noiseno-intrusive-interruptions | issues | high | A fixed cookie banner occupies roughly the bottom 148px of the 800px mobile viewport on load and presents only a prominent Accept action, obscuring app-store/footer content. F-05 high: The on-load cookie banner obscures content and cannot be declined or dismissed. |
maximize-content-reduce-noisesemantic-dismissible-primitives | issues | high | The cookie control is a custom <span role="button" tabindex="0"> inside fixed divs, with no dialog/popover semantics and no light-dismiss or reject action. F-06 medium: Consent uses an ad-hoc pseudo-button and non-semantic overlay. |
maximize-content-reduce-noisereduced-chrome | pass | medium | The unauthenticated pages are visually sparse and devote nearly all non-consent space to login, registration, help, or review content rather than decorative application chrome. |
adapt-to-the-form-factorresponsive-no-horizontal-scroll | pass | high | At 360x800 the layout primitive reported scrollWidth=clientWidth=360 and horizontalOverflowPx=0; mobile screenshots show controls reflowing without clipping. |
adapt-to-the-form-factorcomponent-level-responsiveness | issues | high | CSS inspection found no container queries or logical sizing; at 1365px the login form remains a 244px strip while background panels stretch edge-to-edge, showing no component-level adaptation. F-07 low: Components rely on legacy fixed sizing rather than adapting to their available container. |
adapt-to-the-form-factorinput-modality-aware | issues | high | Every login control, recovery link, and registration field has tabindex=-1, removing the core flows from sequential keyboard navigation; many links are only 14–17px tall and the checkbox is 13x13px. F-08 critical: Core login and registration controls are removed from keyboard navigation and several targets are undersized. |
support-core-task-successclear-purpose-and-primary-action | pass | high | The first viewport identifies the service as mobile dating and exposes sign-in plus a clearly labelled free-registration action; the signup promo has one explicit next step. |
support-core-task-successprimary-flow-completion | issues | high | The login and two-step registration journeys cannot be completed with sequential keyboard input because all controls are tabindex=-1; Turnstile also gates both flows and the invalid registration attempt produced no actionable state. F-09 critical: The primary authentication and registration journeys exclude keyboard users. |
support-core-task-successclear-system-state-and-recovery | issues | high | After accepting consent and attempting an empty registration submission, the page remained unchanged with no :invalid fields, role=alert, aria-invalid, or error text; the 404 tells users to go back but provides no actionable link. F-10 high: Failure states do not provide clear, actionable recovery. |
be-fast-and-stablegood-core-web-vitals | issues | high | Mobile Lighthouse measured LCP 4.3s and FCP 3.7s, outside the good range, although the local desktop trace was faster at 1.35s and CLS remained low. F-11 high: Mobile paint performance is slow despite a visually simple page. |
be-fast-and-stablevisual-stability | pass | high | Observed CLS was 0.037 in Lighthouse and 0.059 in the mobile layout capture, both within the good range; screenshots showed no disruptive content jump. |
be-fast-and-stableefficient-main-thread | pass | high | Lighthouse TBT was 0ms; the trace found one 84.57ms long task and 34.57ms total blocking time, a light main-thread result. |
be-fast-and-stableefficient-resource-delivery | issues | high | Two 300KB+ parser-inserted legacy scripts in <head> have neither async nor defer; HAR recorded 31 requests, seven redirects, and 484KB transferred, with Turnstile alone delaying about 1.39s. F-12 high: Parser-blocking scripts and third-party dependency chains delay a very small page. |
be-fast-and-stabletrim-unused-and-duplicate-code | issues | high | Lighthouse estimates 225KiB of unused JavaScript: 128KB from index.min.js, 54KB from Yandex Metrika, and 48KB from vendor-index.min.js. F-13 high: Most of the shipped JavaScript is unused on the landing page. |
be-inclusivenames-roles-labels | issues | high | axe reported critical label failures for both login fields, an unnamed tracking link, and missing landmarks; registration fields likewise use nearby text rather than associated <label> elements. F-14 critical: Core form fields and some links lack accessible names and structure. |
be-inclusivesufficient-contrast | issues | high | axe found five serious contrast failures, including the 1.98:1 heading, 1.79:1 online count, 3.94:1 sign-in button, 3.26:1 policy link, and 2.49:1 consent button. F-15 high: Important text and controls fail WCAG AA contrast. |
be-inclusivestructure-and-focus | issues | high | axe found no main landmark and content outside landmarks; more critically, login and registration inputs/buttons are all tabindex=-1 despite showing an outline only when focused programmatically. F-16 critical: Landmark structure is weak and the primary flows are absent from keyboard focus order. |
be-inclusivelegible-text | issues | high | The primary h1 is rendered at 13px and the consent text at 12px/14px; screenshots show dense low-contrast copy and edge-to-edge rows that reduce reading comfort. F-17 medium: Key headings and consent copy are cramped and undersized. |
be-inclusivezoom-reflow-targets-and-media | issues | high | Home, signup, registration, recovery, and help omit the viewport meta; the target probe found 13x13 and 14–17px-high controls, while core CAPTCHA canvas interaction has no equivalent exposed in page markup. F-18 high: Most templates omit mobile scaling metadata and expose undersized/keyboard-inaccessible targets. |
follow-best-practicesno-console-errors | pass | high | Lighthouse observed no browser console errors or uncaught exceptions during load. |
follow-best-practicessound-document-and-assets | issues | high | The doctype and charset are valid, but 5 of 6 images omit width/height, four below-fold images are not lazy, and five use legacy PNG/GIF without responsive sources. F-19 medium: Image markup omits intrinsic dimensions and responsive delivery hints. |
follow-best-practicesbrowser-platform-hygiene | issues | high | Lighthouse reports five BFCache blockers (notably Cache-Control: no-store and WebSocket use) plus DevTools cookie issues; deprecated APIs, paste prevention, and on-load permission prompts did pass. F-20 medium: The page cannot use BFCache and generates browser inspector issues. |
be-discoverabletitle-and-description | issues | high | Home has a descriptive title/description, but signup, registration, recovery, and help have no meta description and reviews uses a blank description. F-21 medium: Most public templates lack meaningful meta descriptions. |
be-discoverablecrawlable-and-mobile-friendly | issues | high | Links and robots.txt are crawlable, but four of five representative legacy templates omit meta viewport; several app-store links rely solely on images. F-22 medium: Crawlable pages inconsistently declare mobile behavior and descriptive link text. |
be-discoverablecanonical-and-indexing-signals | issues | medium | Responses and canonicals are valid, but the sitemap index lists only http:// URLs and every lastmod is 2021-09-12 despite currently changing profile/review content. F-23 low: The sitemap exposes stale HTTP-era indexing signals. |
be-discoverablestructured-and-shareable-metadata | issues | high | No JSON-LD or Open Graph tags were present on the organization landing page or public review listing, despite these representing a service/organization and user reviews. F-24 low: Public entity and review pages lack structured and social-preview metadata. |
be-private-and-securesecure-transport-and-headers | issues | high | HTTPS is used, but CSP, HSTS, X-Content-Type-Options, frame protection, Referrer-Policy, and Permissions-Policy are all absent; sid and test cookies lack Secure and SameSite protection. F-25 critical: Core browser security headers and cookie protections are missing. |
be-private-and-securedata-minimisation-and-third-parties | issues | high | While the only consent UI is still visible, Yandex Metrika and LiveInternet load; HAR recorded 25 of 31 requests and 468KB of 484KB from other origins, including repeated Yandex cookie-sync chains. F-26 critical: Tracking begins before meaningful consent and dominates the request surface. |
be-private-and-securein-context-permissions-and-modern-auth | issues | high | Lighthouse confirms no geolocation/notification prompt on load, but both public authentication and registration expose password-only flows; no WebAuthn/passkey capability is present. F-27 medium: Authentication remains password-only with no phishing-resistant option. |
be-private-and-securedefensive-browser-policies | issues | high | Neither CSP frame-ancestors nor X-Frame-Options, HSTS, Referrer-Policy, Permissions-Policy, Trusted Types enforcement, or origin-isolation headers were observed. F-28 high: Browser-enforced defensive policies are absent. |
be-resilientprogressive-enhancement | pass | high | The raw response contains title, h1, description, form content, and core links; discoverability measured 80% content coverage and isJsShell=false, so primary public content survives without JavaScript. |
be-resilientresilient-runtime-behaviour | pass | medium | The audited public templates are server-rendered, use ordinary links/forms, fit at 360px without clipping, and produced no console errors; no fragile menu/overlay positioning was observed beyond separately reported consent semantics. |
be-resilientoffline-and-installable | issues | high | This app-like dating service exposes a manifest, but it contains only name "App" and one 192px icon, with no start_url/display/short_name; no service-worker registrations exist. F-29 low: The app-like service is not meaningfully installable and has no offline fallback. |
be-resilientnetwork-and-http-failure-states | issues | high | The real 404 returns status 404 and explains the problem, but offers only plain-text instructions to go back, no Home/Retry link; no offline fallback is registered. F-30 medium: Network/error recovery surfaces are informational but not actionable. |
be-internationalisedlang-dir-and-logical-properties | issues | high | Home, recovery, reviews, and registration use xml:lang or no language instead of a valid HTML lang attribute; axe flags html-has-lang, and the CSS capability probe found no logical properties. F-31 high: Most templates do not expose a valid document language and use physical CSS. |
be-internationalisedlocale-aware-data | issues | medium | The reviews template renders Russian relative dates such as “вчера в 21:54” in <time> elements without datetime values, leaving no machine-readable instant and no evidence of locale-aware formatting. F-32 low: Human-readable review times omit machine-readable locale-neutral values. |
be-internationalisedtime-zone-correctness | not-applicable | high | The audited public/auth/help templates expose no scheduled event, recurring interval, absolute appointment time, or other user-editable time-zone-sensitive concept. |
be-trustworthyno-dark-patterns | issues | high | The consent banner offers only Accept while analytics and cookie syncing already run; there is no Reject or Settings control, making the apparent choice non-reversible and non-neutral. F-33 critical: Consent is one-sided and tracking begins before the user can choose. |
be-trustworthyhumane-error-handling | issues | high | After accepting terms and attempting an empty registration submission, no visible error, :invalid field, aria-invalid state, role=alert, or focused recovery target appeared; the CAPTCHA remained in “Verifying...” state. F-34 high: Invalid registration fails silently instead of explaining what to correct. |
be-trustworthytrustworthy-input-assistance | issues | high | Login and registration text/password/date fields have empty autocomplete values and tabindex=-1; fields rely on nearby text rather than explicit labels, preventing autofill and keyboard assistance. F-35 high: Authentication fields do not support standard autofill or keyboard input order. |
be-trustworthysafe-commercial-and-account-flows | issues | high | The audited account-creation/sign-in flows are password-only, keyboard-inaccessible, gated by a canvas CAPTCHA, and combine terms, cookies, and personal-data consent into one required checkbox. F-36 high: Account onboarding bundles consent and excludes safer, accessible authentication choices. |
be-sustainableoptimised-assets | issues | high | Five images use legacy PNG/GIF, five omit intrinsic dimensions, four below-fold images are not lazy, and no image uses srcset; transferred image weight is modest at 36KB but delivery is not optimized. F-37 low: Image delivery omits modern sizing and loading optimizations. |
be-sustainableno-wasteful-work | issues | high | A simple server-rendered login page transfers 428KB of script and Lighthouse estimates 225KiB unused; analytics triggers multiple cookie-sync redirects before consent. F-38 high: Unused code and pre-consent analytics consume resources without immediate user value. |
be-sustainablethird-party-and-media-budget | issues | high | 25 of 31 requests and 468KB of 484KB transferred came from other origins; Yandex, Cloudflare, and LiveInternet account for substantial script/network activity on a minimal login view. F-39 high: Third-party work dominates the page’s network budget. |
be-agent-readystructured-agent-capabilities | not-applicable | high | No developer intent or public agent-facing workflow was declared; this emerging capability is not required for the audited conventional dating/auth site. |
be-agent-readyon-device-inference | not-applicable | high | No summarisation, generation, translation, or other inference task is part of the audited public login/registration/help journeys. |
be-memory-efficientno-leak-under-repeated-interaction | pass | medium | After ten representative login input/focus cycles, heap self-size rose only 59,176 bytes (6,693,096 to 6,752,272; 0.9%), node count by 142, while closure count stayed exactly 8,393. |
be-memory-efficientbounded-footprint | pass | medium | The loaded page heap summary was about 6.7MB with stable DOM element count of 92, proportionate for the login page and its CAPTCHA/analytics dependencies. |
be-memory-efficientno-detached-dom-or-unbounded-listeners | pass | medium | Across the repeated interaction, native node population remained effectively flat (18,967 to 18,968), closures stayed 8,393, DOM elements stayed 92, and no Detached* constructor appeared in either top-constructor summary. |
Provenance
Canonical report: results/atomic/reports/0849-m_tabor_ru.json
Report SHA-256: 70dd97d044232690faea48bc8333f94e629d4a1d786c09423e8e779f9ffd3cdf
Local retained report: runs/2026-07-17T17-27-24-856Z/atomic-reports/m_tabor_ru/2026-07-23T10-03-16-134Z/report.json
Local evidence root: runs/2026-07-17T17-27-24-856Z/atomic-reports/m_tabor_ru/2026-07-23T10-03-16-134Z
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.