Manifest position 793 · CrUX rank bucket 1000
https://tenki.jp
Coverage complete
Coverage-complete audit of eight representative paths/templates. Excluded: destructive account actions, authenticated member pages and every regional URL; sampled templates represent those repeated surfaces.
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; computed styles report color-scheme: normal and a white body background under prefers-color-scheme: dark. F01 medium: The site ignores the user’s dark colour-scheme preference. |
respect-user-preferencesrespects-reduced-motion | pass | high | Reduced-motion emulation found zero active animations and no auto-advancing content. |
respect-user-preferencesrespects-contrast | pass | high | Forced-colors/prefers-contrast capture preserved visible text, controls, map labels and focus affordances. |
implement-natural-interactionsview-transitions | issues | medium | CSS inspection found no view-transition rules while the homepage exposes eleven JavaScript date-state links that replace forecast state. F02 low: Interactive date/state changes are abrupt and do not use View Transitions. |
implement-natural-interactionsscroll-driven-animations | not-applicable | high | No scroll-linked animation or scrollytelling surface was present; native scrolling has no scroll-event animation to implement. |
implement-natural-interactionsphysical-gestures | pass | high | Representative pages use native scrolling and ordinary links/controls; no custom pointer-drag gesture that fights platform scrolling was observed. |
provide-guided-navigationscroll-state-aware-chrome | issues | medium | The 7,062 CSS-pixel article and other long templates retain static top navigation; CSS inspection found no scroll-state query, progress indicator or scroll timeline. F03 low: Long pages provide static chrome rather than scroll-state guidance. |
provide-guided-navigationanchored-positioning | not-applicable | high | No tooltip, transient popover, or edge-positioned menu was present on the representative public paths. |
provide-guided-navigationdirects-attention | pass | high | Breadcrumbs, selected blue navigation tabs, section headings and search-result rows consistently show location and next destination. |
maximize-content-reduce-noiseno-intrusive-interruptions | pass | high | Initial screenshots on all representative paths show no modal, consent wall or viewport-covering interstitial; content is immediately present. |
maximize-content-reduce-noisesemantic-dismissible-primitives | not-applicable | high | No modal/dialog/disclosure flow or transient overlay was present on load or in the weather lookup journey. |
maximize-content-reduce-noisereduced-chrome | issues | high | Desktop and article screenshots show background advertising, a right ad rail and dense repeated navigation around the weather/article content; the HAR records extensive ad-tech delivery. F04 medium: Advertising and duplicated navigation substantially crowd the primary weather content. |
adapt-to-the-form-factorresponsive-no-horizontal-scroll | issues | high | Every tested 360x800 page lacked a viewport meta tag and rendered a 1088–1248 CSS-pixel desktop canvas scaled to 0.29–0.33. Text and controls are miniature across homepage, forecast, radar, earthquake, article and search. F05 critical: Public pages ship a fixed desktop layout to phones, making core weather tasks tiny and difficult to operate. |
adapt-to-the-form-factorcomponent-level-responsiveness | issues | high | Every tested 360x800 page lacked a viewport meta tag and rendered a 1088–1248 CSS-pixel desktop canvas scaled to 0.29–0.33. Text and controls are miniature across homepage, forecast, radar, earthquake, article and search. F05 critical: Public pages ship a fixed desktop layout to phones, making core weather tasks tiny and difficult to operate. |
adapt-to-the-form-factorinput-modality-aware | issues | high | Every tested 360x800 page lacked a viewport meta tag and rendered a 1088–1248 CSS-pixel desktop canvas scaled to 0.29–0.33. Text and controls are miniature across homepage, forecast, radar, earthquake, article and search. F05 critical: Public pages ship a fixed desktop layout to phones, making core weather tasks tiny and difficult to operate. |
support-core-task-successclear-purpose-and-primary-action | pass | high | The tenki.jp identity, address search and national forecast are immediately visible, and emergency/radar routes have explicit labels. |
support-core-task-successprimary-flow-completion | issues | high | Every tested 360x800 page lacked a viewport meta tag and rendered a 1088–1248 CSS-pixel desktop canvas scaled to 0.29–0.33. Text and controls are miniature across homepage, forecast, radar, earthquake, article and search. F05 critical: Public pages ship a fixed desktop layout to phones, making core weather tasks tiny and difficult to operate. |
support-core-task-successclear-system-state-and-recovery | pass | high | A nonexistent route returns HTTP 404 with a clear Japanese explanation and prominent TOPへ戻る recovery link; search results are explicit rows. |
be-fast-and-stablegood-core-web-vitals | issues | high | Lighthouse measured LCP 4.64s, TBT 770ms, Speed Index 16.1s and performance 0.47; the independent trace measured LCP 12.50s and four long tasks. F06 high: The homepage misses fast-load and interaction performance targets. |
be-fast-and-stablevisual-stability | pass | high | Observed CLS was 0.009 desktop and 0.018 in mobile emulation, with all representative templates below 0.02. |
be-fast-and-stableefficient-main-thread | issues | high | Lighthouse measured LCP 4.64s, TBT 770ms, Speed Index 16.1s and performance 0.47; the independent trace measured LCP 12.50s and four long tasks. F06 high: The homepage misses fast-load and interaction performance targets. |
be-fast-and-stableefficient-resource-delivery | issues | high | HAR: 356 requests and 4.90 MB transferred, including 72 scripts/2.71 MB and 343 third-party requests. Lighthouse estimates 1,562 KiB unused JavaScript and reports render-blocking savings of 2,980ms. F07 high: Third-party and unused code dominate resource delivery. |
be-fast-and-stabletrim-unused-and-duplicate-code | issues | high | HAR: 356 requests and 4.90 MB transferred, including 72 scripts/2.71 MB and 343 third-party requests. Lighthouse estimates 1,562 KiB unused JavaScript and reports render-blocking savings of 2,980ms. F07 high: Third-party and unused code dominate resource delivery. |
be-inclusivenames-roles-labels | issues | high | axe found critical image-alt failures on three images, one unnamed member link, and Lighthouse found an unnamed button; the address input has no programmatic label. F08 high: Interactive and image content lacks accessible names. |
be-inclusivesufficient-contrast | issues | high | axe reports 44 serious colour-contrast failures and links distinguishable only by colour in text blocks. F09 high: Text and links fail minimum contrast requirements. |
be-inclusivestructure-and-focus | issues | high | axe found no main landmark, 63 regions outside landmarks, duplicate navigation landmarks, skipped heading order and tabindex=2. The DOM contains an H1 followed by H3 sections. F10 high: Landmarks, headings and focus order are structurally inconsistent. |
be-inclusivelegible-text | issues | high | Every tested 360x800 page lacked a viewport meta tag and rendered a 1088–1248 CSS-pixel desktop canvas scaled to 0.29–0.33. Text and controls are miniature across homepage, forecast, radar, earthquake, article and search. F05 critical: Public pages ship a fixed desktop layout to phones, making core weather tasks tiny and difficult to operate. |
be-inclusivezoom-reflow-targets-and-media | issues | high | Every tested 360x800 page lacked a viewport meta tag and rendered a 1088–1248 CSS-pixel desktop canvas scaled to 0.29–0.33. Text and controls are miniature across homepage, forecast, radar, earthquake, article and search. F05 critical: Public pages ship a fixed desktop layout to phones, making core weather tasks tiny and difficult to operate. |
follow-best-practicesno-console-errors | issues | high | Lighthouse errors-in-console audit failed during a clean homepage load. F11 medium: The page logs runtime errors. |
follow-best-practicessound-document-and-assets | issues | high | The images audit found 9 images without dimensions, 10 oversized images and 60 of 66 in legacy formats; doctype and UTF-8 charset were valid. F12 medium: Document assets are incompletely sized and inefficiently encoded. |
follow-best-practicesbrowser-platform-hygiene | issues | high | Lighthouse reports two deprecated-API warnings and a failed inspector-issues audit. F13 medium: Deprecated APIs and DevTools inspector issues remain in production. |
be-discoverabletitle-and-description | pass | high | Homepage and article have unique descriptive titles and meta descriptions matching visible content. |
be-discoverablecrawlable-and-mobile-friendly | issues | high | The page has no viewport meta tag and 17 javascript: links; Lighthouse fails crawlable anchors and mobile viewport audits. robots.txt itself is reachable and advertises two sitemap URLs. F14 high: Mobile and crawler navigation signals contain avoidable defects. |
be-discoverablecanonical-and-indexing-signals | pass | high | Pages returned 200 with self-canonical URLs; robots.txt allows public paths and advertises sitemap_pc.xml and sitemap_lite.xml; 404 pages return 404/noindex. |
be-discoverablestructured-and-shareable-metadata | pass | high | The article supplies NewsArticle and BreadcrumbList JSON-LD plus complete Open Graph title, description, URL, image and ja_JP locale. |
be-private-and-securesecure-transport-and-headers | issues | high | HTTPS, HSTS, nosniff and frame protection are present, but CSP is absent. Cookie inspection found 17 cookies without Secure and many SameSite=None/long-lived analytics identifiers. F15 high: Security policy and cookie defaults are too weak. |
be-private-and-securedata-minimisation-and-third-parties | issues | high | Tracker capture found 70 third-party origins and 26 known tracker domains. HAR attributes 343/356 requests and 4.86/4.90 MB to non-main origins, including DoubleClick, Taboola, Meta, TikTok and Clarity. F16 high: The default page load has an excessive tracking and third-party footprint. |
be-private-and-securein-context-permissions-and-modern-auth | issues | high | The login page exposes email and password fields, no passkey/WebAuthn affordance, and the email field has an empty autocomplete token; the OAuth request does use PKCE. F17 medium: Authentication is password-only and misses complete autofill semantics. |
be-private-and-securedefensive-browser-policies | issues | high | Permissions-Policy is missing and HSTS max-age is only 300 seconds; Referrer-Policy and X-Frame-Options are present. F18 medium: Defensive browser policy coverage is incomplete. |
be-resilientprogressive-enhancement | pass | high | No-JavaScript crawler screenshots retain the forecast shell and the article headline, author, publication time and breadcrumbs; the site is not an empty JS shell. |
be-resilientresilient-runtime-behaviour | pass | high | Weather, radar and article templates loaded to stable content; date-tab interaction completed ten repetitions with unchanged DOM node count. |
be-resilientoffline-and-installable | not-applicable | high | Live weather, radar and warning data are intrinsically network-current; offline/installability is not a necessary promise for the audited public website. |
be-resilientnetwork-and-http-failure-states | pass | high | The tested invalid URL returns a real 404 page with explanation and a clear route back to the homepage rather than a blank shell. |
be-internationalisedlang-dir-and-logical-properties | pass | high | The document declares lang=ja, Japanese reading order is correct, and stylesheet inspection found logical inline properties. |
be-internationalisedlocale-aware-data | pass | high | Dates, weekday labels, temperatures and publication times are rendered in appropriate Japanese forms throughout forecast/article templates. |
be-internationalisedtime-zone-correctness | pass | high | Article timestamps use explicit +09:00 offsets and the UI identifies publication/display time in Japanese local context. |
be-trustworthyno-dark-patterns | pass | high | No consent wall, confirmshaming, forced continuity or preselected commercial commitment appeared in the representative weather and login paths. |
be-trustworthyhumane-error-handling | pass | high | The public search is optional free text with predictable GET results; the login page labels fields and provides a visible password-recovery route rather than premature errors. |
be-trustworthytrustworthy-input-assistance | issues | high | The address/postcode search input has no autocomplete token or label; the login email field also has empty autocomplete while only the password has current-password. F19 medium: Search and login inputs do not provide complete autocomplete assistance. |
be-trustworthysafe-commercial-and-account-flows | pass | high | The optional login clearly identifies email/password, links password recovery and terms, and uses an OAuth authorization request with PKCE; no purchase/subscription commitment is present. |
be-sustainableoptimised-assets | issues | high | Images audit: 60/66 legacy-format images, 18 without srcset, 10 oversized and 32 below-fold images not lazy-loaded; one decorative banner transfers 876 KB. F20 medium: Images are not consistently optimised for transfer or display size. |
be-sustainableno-wasteful-work | issues | high | A single load triggers 134 XHR/fetch requests and 70 third-party origins; Lighthouse records 9.5s main-thread work and 7.0s JavaScript execution. F21 high: Background ad and analytics work is disproportionate to the weather task. |
be-sustainablethird-party-and-media-budget | issues | high | Third-party delivery accounts for 4.86 MB and 343 requests; large ad/recommendation scripts and decorative images surround a comparatively lightweight forecast document. F22 high: The third-party and media budget is not proportionate. |
be-agent-readystructured-agent-capabilities | not-applicable | high | This is an informational weather publisher with crawlable pages and structured metadata, not a transactional tool surface that needs WebMCP actions. |
be-agent-readyon-device-inference | not-applicable | high | No summarisation or generative task is part of the core weather lookup; on-device inference would add no necessary user value. |
be-memory-efficientno-leak-under-repeated-interaction | pass | high | Ten forecast date-tab activations raised sampled used heap about 1.5 MB but left live DOM elements unchanged at 2,320; separate heap summaries differed about 4.1%, insufficient to establish unbounded retention. |
be-memory-efficientbounded-footprint | issues | high | Baseline heap summary contains 1,456,317 nodes and 73.7 MB self size, including 161,933 closures. After date-tab repetition it reaches 76.7 MB, although DOM count remained stable and this pair did not prove an unbounded leak. F23 medium: The homepage has a disproportionately large initial memory footprint. |
be-memory-efficientno-detached-dom-or-unbounded-listeners | pass | high | Neither compact heap summary surfaced Detached* among top constructors, and repeated date-tab interaction left DOM count unchanged. |
Provenance
Canonical report: results/atomic/reports/0793-tenki_jp.json
Report SHA-256: 53d12465886571a3e68cee6a8b447df59e5aa18eeb263cb1a8ea8c72a7b50bf1
Local retained report: runs/2026-07-17T17-27-24-856Z/atomic-reports/tenki_jp/2026-07-23T01-50-30-795Z/report.json
Local evidence root: runs/2026-07-17T17-27-24-856Z/atomic-reports/tenki_jp/2026-07-23T01-50-30-795Z
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.