Manifest position 555 · CrUX rank bucket 1000
https://uranai.nosv.org
Coverage complete
Coverage-complete agentic audit of four representative templates plus a not-found recovery state.
All 58 atomic check outcomes
| Principle / check | Status | Confidence | Evidence or reason |
|---|---|---|---|
respect-user-preferencesrespects-color-scheme | issues | high | The dark/reduced-motion probe matched dark mode but body colors remained the light palette (rgb(179, 213, 13), black text); the dark screenshot is byte-identical to the default capture. F01 medium: The site ignores the user’s dark-theme preference. |
respect-user-preferencesrespects-reduced-motion | pass | high | Emulated prefers-reduced-motion: reduce matched and document.getAnimations() returned zero animations. |
respect-user-preferencesrespects-contrast | pass | high | Forced-colors and prefers-contrast emulation produced black/white system colors with visible bordered controls. |
implement-natural-interactionsview-transitions | issues | high | The DOM/CSS probe found no view-transition CSS or startViewTransition usage across the MPA entry surface. F02 medium: Page and state changes have no View Transition treatment. |
implement-natural-interactionsscroll-driven-animations | not-applicable | high | No scroll-linked animation exists on the representative templates, so there is no implementation to evaluate. |
implement-natural-interactionsphysical-gestures | not-applicable | high | The representative MPA pages expose no swipe, pull, carousel, or gesture-driven interaction. |
provide-guided-navigationscroll-state-aware-chrome | issues | high | The home page is over 2,100 CSS px tall, while CSS inspection found no scroll-state container rules, progress indicator, or animation timeline. F03 low: Long pages provide no scroll-state-aware navigation or progress feedback. |
provide-guided-navigationanchored-positioning | not-applicable | high | No site-owned tooltip, popover, or anchored menu was present on the representative paths. |
provide-guided-navigationdirects-attention | issues | high | The DOM contains hundreds of hash-only links and no View Transition, highlight, or scroll-marker signal; the crawler screenshot also shows overlapping in-page navigation over content. F04 medium: In-page navigation does not clearly cue the destination or focus change. |
maximize-content-reduce-noiseno-intrusive-interruptions | issues | high | Desktop, mobile, detail, listing, and creation screenshots all open behind a modal consent wall for personalised advertising and 210 partners. F05 high: A consent interstitial obscures the entire primary experience on every fresh visit. |
maximize-content-reduce-noisesemantic-dismissible-primitives | issues | high | The probe found four role=dialog overlays but zero dialog or popover elements; the first layer offers only Consent and Manage options. F06 medium: Consent UI uses custom role dialogs rather than native dialog semantics and has no direct dismiss/decline action. |
maximize-content-reduce-noisereduced-chrome | issues | high | The desktop/crawler screenshots show multiple stacked navigation strips, boxed regions, sidebars, icon rows, and overlapping in-page navigation competing with the main content. F07 medium: Dense legacy chrome and repeated sidebars crowd the content. |
adapt-to-the-form-factorresponsive-no-horizontal-scroll | issues | high | At a requested 360x800 viewport, layout reported no viewport meta and a 980px layout/client width at scale 0.367; text and controls are tiny in the mobile screenshot. F08 critical: Mobile devices receive a 980px desktop layout scaled down to fit. |
adapt-to-the-form-factorcomponent-level-responsiveness | issues | high | CSS inspection across the representative page found zero @container rules; the whole desktop composition is merely scaled on mobile. F09 medium: Components do not adapt to their available container. |
adapt-to-the-form-factorinput-modality-aware | issues | high | Programmatic focus on the first link computed outline: none 0px and no box shadow; the absent viewport meta shrinks controls to 36.7% on mobile. F10 high: Keyboard focus is not visibly styled and scaled mobile targets are too small. |
support-core-task-successclear-purpose-and-primary-action | issues | high | Every fresh-load screenshot places the advertising-consent decision above the site’s search, read, and create actions. F11 high: The consent wall hides the site purpose and primary actions on arrival. |
support-core-task-successprimary-flow-completion | issues | high | The homepage, sample story, listing, and 329-control creation form are all blocked visually until the visitor opens consent options or accepts. F12 high: Reading and creation journeys begin with an avoidable consent detour. |
support-core-task-successclear-system-state-and-recovery | pass | high | The tested unknown URL returns a concise Japanese not-found state with a working “back to top” recovery link; the create page also explains editing and recovery constraints. |
be-fast-and-stablegood-core-web-vitals | issues | high | Lighthouse measured FCP 4.5s, LCP 5.7s, Speed Index 7.4s, and performance 0.63; the independent trace measured LCP 3.77s. F13 high: Largest Contentful Paint is well outside the good range. |
be-fast-and-stablevisual-stability | pass | high | Layout observers measured CLS 0.00095 on home and 0 on detail; images with dimensions reserve the main hero area. |
be-fast-and-stableefficient-main-thread | pass | high | Trace and layout evidence recorded zero long tasks and 0ms total blocking time during load. |
be-fast-and-stableefficient-resource-delivery | issues | high | HAR recorded 94 requests and identified parser-inserted jQuery 1.4.4, jquery-cookie, localStorage, analytics, and other scripts; DOM confirmation shows these scripts are neither async nor defer. F14 high: Legacy synchronous scripts and dependency chains delay rendering. |
be-fast-and-stabletrim-unused-and-duplicate-code | issues | high | HAR measured 633,623 script bytes; ads (190KB), GTM (158KB), and consent scripts (about 144KB) are the largest resources, with 92 of 94 requests classed third-party. F15 high: Third-party JavaScript dominates the transfer budget. |
be-inclusivenames-roles-labels | issues | high | The image audit found 40 of 44 images without alt. Lighthouse failed image-alt, link-name, and frame-title. The creation form has 179 visible non-button controls without labels. F16 critical: Core images, links, frames, and creation inputs lack accessible names. |
be-inclusivesufficient-contrast | issues | high | Lighthouse’s color-contrast audit failed, and its link-in-text-block audit found links relying on color alone. F17 high: Text and essential links fail minimum contrast. |
be-inclusivestructure-and-focus | issues | high | There is no main/nav/header/footer landmark, Lighthouse failed heading order and main-landmark checks, and focused links compute no outline. F18 high: Document structure and keyboard focus are not reliable. |
be-inclusivelegible-text | issues | high | At 360px, Chrome lays out 980 CSS px and scales the page to 0.367, making body copy and controls too small to read comfortably. F19 high: Mobile text is rendered as a miniaturised desktop page. |
be-inclusivezoom-reflow-targets-and-media | issues | high | Lighthouse failed meta-viewport; both homepage and detail layout captures remained 980px wide at a requested 360px viewport. F20 high: The page does not reflow for a narrow viewport. |
follow-best-practicesno-console-errors | pass | high | Lighthouse errors-in-console audit passed. |
follow-best-practicessound-document-and-assets | issues | high | Lighthouse reported a doctype failure; 38 of 44 images lack dimensions, 43 use legacy formats, and 34 below-fold images are not lazy-loaded. F21 medium: Document and image asset hygiene is poor. |
follow-best-practicesbrowser-platform-hygiene | pass | high | Lighthouse deprecation and BFCache audits passed, and the probe found no permission requests on load. |
be-discoverabletitle-and-description | pass | high | All four representative DOM captures have descriptive Japanese titles; the homepage has a substantive meta description. |
be-discoverablecrawlable-and-mobile-friendly | issues | high | All four templates lack a viewport meta tag; mobile evidence renders a 980px page at 36.7% scale. Robots.txt is reachable and Lighthouse says anchors are crawlable. F22 high: The public site is not mobile-friendly despite crawlable links. |
be-discoverablecanonical-and-indexing-signals | issues | high | The home probe found no canonical link; /sitemap.xml returns 404. The discoverability fetch also reported an unexpected finalUrl using http although the browser remained on HTTPS. F23 medium: Canonical and sitemap signals are incomplete. |
be-discoverablestructured-and-shareable-metadata | issues | high | The homepage has Open Graph tags but zero JSON-LD blocks; the sample content page represents a creative work that could expose schema.org metadata. F24 medium: Entity pages lack structured data. |
be-private-and-securesecure-transport-and-headers | issues | high | Headers found no CSP, HSTS, or nosniff. All 12 observed cookies lack Secure and use SameSite=None, including several 390–400 day cookies. F25 critical: HTTPS is undermined by missing security headers and insecure cookies. |
be-private-and-securedata-minimisation-and-third-parties | issues | high | Tracker evidence found 11 third-party origins and four known tracker domains; the consent UI names 210 partners and the HAR assigns 859,735 of 879,977 transferred bytes to third parties. F26 high: The page has an excessive tracking and third-party footprint. |
be-private-and-securein-context-permissions-and-modern-auth | issues | high | The creation flow exposes a password field and current-page probes found no WebAuthn/PublicKeyCredential usage; no permission calls fire on load. F27 medium: Account and anonymous-edit flows rely on passwords without phishing-resistant authentication. |
be-private-and-securedefensive-browser-policies | issues | high | Headers found no clickjacking protection, Referrer-Policy, Permissions-Policy, CSP/frame-ancestors, HSTS, or nosniff. F28 high: Browser-enforced defensive policies are absent. |
be-resilientprogressive-enhancement | pass | high | Crawler screenshots with JavaScript disabled show server-rendered homepage and story content; neither path is a JS shell. |
be-resilientresilient-runtime-behaviour | pass | high | All four server-rendered templates loaded to usable DOM states, and the consent overlay remained within the captured viewport under tested sizes. |
be-resilientoffline-and-installable | issues | high | /manifest.json exists, but the DOM has no manifest link and there is no controlling service worker. F29 medium: The app-like creation service is neither linked as installable nor available offline. |
be-resilientnetwork-and-http-failure-states | pass | high | A representative missing URL returned an explicit not-found message and a working recovery link instead of a blank shell. |
be-internationalisedlang-dir-and-logical-properties | not-applicable | high | This is a deliberately Japanese-language service; no alternate-language or bidi surface was found in scope. |
be-internationalisedlocale-aware-data | not-applicable | high | No locale-switching or user-facing currency/date formatting flow was present in the representative scope. |
be-internationalisedtime-zone-correctness | not-applicable | high | No event scheduling, time-zone, or DST-sensitive surface was present in the representative scope. |
be-trustworthyno-dark-patterns | issues | high | The first layer presents a prominent Consent button and Manage options, but no Reject/Decline button, while requesting personalised processing by 210 partners. F30 high: The consent choice is asymmetric and a direct refusal is hidden. |
be-trustworthyhumane-error-handling | issues | high | The 329-control form has zero required controls and 179 visible unlabelled controls; this prevents dependable native validation and accessible error association. F31 high: The creation form lacks robust field semantics for preventing and explaining mistakes. |
be-trustworthytrustworthy-input-assistance | issues | high | Only one of 267 visible controls has a non-empty autocomplete value; email and password/account-adjacent inputs do not expose appropriate tokens in the captured DOM. F32 medium: The large creation form provides almost no autocomplete assistance. |
be-trustworthysafe-commercial-and-account-flows | issues | high | The create flow asks for email and an editing password while the site offers no passkey signal; users are warned that guessed passwords permit unauthorised editing. F33 high: Password-based anonymous editing creates avoidable account and recovery risk. |
be-sustainableoptimised-assets | issues | high | The image audit found 43 legacy-format images, zero modern-format images, eight missing srcset, and 34 below-fold images without lazy loading. F34 medium: Most images use legacy formats and lack responsive delivery. |
be-sustainableno-wasteful-work | issues | high | HAR records 92 third-party requests and 633KB of scripts for a text/listing page; tracker evidence confirms active analytics and ad domains. F35 high: Optional advertising and analytics create disproportionate background work. |
be-sustainablethird-party-and-media-budget | issues | high | 859,735 of 879,977 transferred bytes are attributed to third-party origins; ads, tag management, consent, and a 48KB font dominate a largely text-based page. F36 high: Third-party resources consume nearly the entire page budget. |
be-agent-readystructured-agent-capabilities | not-applicable | high | No declared agent-facing intent or transactional agent surface was found; this emerging capability is out of scope. |
be-agent-readyon-device-inference | not-applicable | high | No user task in the representative scope requires or clearly benefits from on-device inference. |
be-memory-efficientno-leak-under-repeated-interaction | not-applicable | high | The audited site uses full-document navigation and no safe, bounded, representative in-page interaction suitable for a 10x repeat was identified; no synthetic interaction was fabricated. |
be-memory-efficientbounded-footprint | pass | high | The heap summary measured about 18.2MB self size for the content page plus consent UI, proportionate to the captured third-party-heavy state. |
be-memory-efficientno-detached-dom-or-unbounded-listeners | pass | medium | The heap constructor summary exposed no Detached* constructor among retained top constructors in the single tested state; no repeated interaction was available. |
Provenance
Canonical report: results/atomic/reports/0555-uranai_nosv_org.json
Report SHA-256: 70016311a6f5c76f144b22dbaf50a107403c69f392fde3c425817f4a49ffdf68
Local retained report: runs/2026-07-17T17-27-24-856Z/atomic-reports/uranai_nosv_org/2026-07-21T13-27-21-313Z/report.json
Local evidence root: runs/2026-07-17T17-27-24-856Z/atomic-reports/uranai_nosv_org/2026-07-21T13-27-21-313Z
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.