Manifest position 251 · CrUX rank bucket 1000
https://m.search.naver.com
Coverage complete
Coverage-complete audit of the mobile entry, search-result, and error-recovery templates. Not covered: logged-in account, shopping/checkout, media playback, and every vertical tab because they are separate products/templates requiring authenticated or domain-specific runs.
All 58 atomic check outcomes
| Principle / check | Status | Confidence | Evidence or reason |
|---|---|---|---|
respect-user-preferencesrespects-color-scheme | issues | high | The dark-mode screenshot remains a light gray/white interface and the page probe reports computed colorScheme “normal”. F01 medium: Dark preference does not retint the interface |
respect-user-preferencesrespects-reduced-motion | issues | high | Under `prefers-reduced-motion: reduce`, four animations were still running (300–400 ms). F02 medium: Motion continues when reduced motion is requested |
respect-user-preferencesrespects-contrast | issues | high | Lighthouse found 3.36:1 tab text contrast and the page exposes no `color-scheme`; no forced-colors adaptation was evident in captured CSS. F03 medium: Contrast preferences are not fully supported |
implement-natural-interactionsview-transitions | issues | high | DOM/CSS inspection found no view-transition declarations although the search/result state changes substantially. F04 low: Primary state changes do not use View Transitions |
implement-natural-interactionsscroll-driven-animations | not-applicable | high | No parallax, scrollytelling, or entry/exit scroll-linked animation was observed on the representative search templates. |
implement-natural-interactionsphysical-gestures | pass | medium | The mobile result tabs and content carousels use horizontal overflow/flick structures without blocking page scroll; no broken gesture was observed. |
provide-guided-navigationscroll-state-aware-chrome | pass | medium | The result page keeps its search/category chrome at the top and preserves query context while the 6,831 px page scrolls. |
provide-guided-navigationanchored-positioning | not-applicable | high | No tooltip/popover positioned near a viewport edge was required or opened in the covered paths. |
provide-guided-navigationdirects-attention | pass | medium | The query remains visible, the selected category is visually grouped, skeletons communicate pending content, and result sections follow reading order. |
maximize-content-reduce-noiseno-intrusive-interruptions | pass | medium | Mobile screenshots show content immediately with no full-viewport consent wall or load-time interstitial. |
maximize-content-reduce-noisesemantic-dismissible-primitives | issues | high | The results DOM contains multiple custom “layer” controls but zero `<dialog>`, `[popover]`, or `<details>` elements. F05 medium: Autocomplete and information layers use custom overlay structures |
maximize-content-reduce-noisereduced-chrome | pass | medium | The search-results first viewport prioritizes query, categories, AI status and results without a desktop frame or excessive borders. |
adapt-to-the-form-factorresponsive-no-horizontal-scroll | pass | medium | At 360×800 both tested templates reported scrollWidth=clientWidth=360 and viewport metadata was present. |
adapt-to-the-form-factorcomponent-level-responsiveness | issues | high | The stylesheet probe counted zero container-query rules across the rendered result page. F06 low: Reusable components do not show container-level adaptation |
adapt-to-the-form-factorinput-modality-aware | issues | high | The probe found 69 targets below 24 px in one dimension, including 3×16 px menu buttons; focus inspection found many links with `outline: none`, and Lighthouse failed target-size. F07 high: Touch and keyboard affordances are too small or invisible |
support-core-task-successclear-purpose-and-primary-action | pass | medium | The entry search field is prominent and the results title/query make the page purpose and next action immediately clear. |
support-core-task-successprimary-flow-completion | pass | medium | A direct query produced a populated results page with 316 crawlable links and relevant result modules. |
support-core-task-successclear-system-state-and-recovery | pass | medium | The results screenshot shows loading skeletons, while the invalid route gives a clear explanation plus Previous page and Naver home recovery actions. |
be-fast-and-stablegood-core-web-vitals | issues | high | Lighthouse measured LCP 34.2 s, TBT 840 ms and CLS 0.166; the independent layout run measured CLS 0.327. F08 high: Mobile search results miss the Core Web Vitals performance bar |
be-fast-and-stablevisual-stability | issues | high | The mobile layout observer recorded CLS 0.327 across many shifts while skeleton/result content populated. F09 high: Search results shift substantially during load |
be-fast-and-stableefficient-main-thread | issues | high | Lighthouse measured 840 ms TBT; the layout observer saw a 236 ms long task and the trace saw three long tasks. F10 high: Heavy scripting blocks the main thread |
be-fast-and-stableefficient-resource-delivery | issues | high | HAR captured 119 requests and 5.68 MB, including 1.62 MB script, 774 KB fonts, and parser-inserted high-priority CSS/script candidates. F11 high: The initial network payload is large and contains blocking dependencies |
be-fast-and-stabletrim-unused-and-duplicate-code | issues | high | The HAR identifies a 914 KB vendor script and 238 KB main script; Lighthouse reports substantial unused JavaScript/CSS opportunities. F12 high: Large JavaScript payloads dominate loading |
be-inclusivenames-roles-labels | issues | high | Lighthouse failed ARIA allowed attributes, command names, link names, and image alt; the image probe found 33 of 50 images without alt. F13 high: Interactive controls and images lack accessible semantics |
be-inclusivesufficient-contrast | issues | high | Lighthouse measured 3.36:1 for 16 px search tabs against white, below the required 4.5:1. F14 high: Result navigation text fails minimum contrast |
be-inclusivestructure-and-focus | issues | high | Lighthouse reports no main landmark; the focus probe found numerous focusable links with no outline and an unlabeled text search input. F15 high: Landmark and focus behavior is inconsistent |
be-inclusivelegible-text | pass | medium | The mobile screenshots show readable Korean/Latin text, fluid card wrapping, and no clipping or horizontal overflow. |
be-inclusivezoom-reflow-targets-and-media | issues | high | Although user scaling is allowed and no horizontal overflow occurred, Lighthouse failed target-size and the probe counted 69 undersized targets. F16 high: Small targets undermine zoomed and touch use |
follow-best-practicesno-console-errors | pass | medium | Lighthouse `errors-in-console` passed for the audited results load. |
follow-best-practicessound-document-and-assets | issues | high | The image audit found 23 images without width/height, 25 without responsive sources, and 5 legacy-format images. F17 medium: Image asset markup is incomplete |
follow-best-practicesbrowser-platform-hygiene | issues | high | Lighthouse reported one deprecated-API warning on the results page. F18 medium: Deprecated browser API usage remains |
be-discoverabletitle-and-description | pass | medium | The results page has title “web platform : 네이버 검색” and a query-specific Korean meta description. |
be-discoverablecrawlable-and-mobile-friendly | issues | high | Lighthouse reports the results page blocked from indexing and four links with non-descriptive text, although all 316 anchors had hrefs and viewport metadata was present. F19 medium: Indexing and link semantics are inconsistent |
be-discoverablecanonical-and-indexing-signals | issues | high | The results probe found no canonical or hreflang links, Lighthouse says indexing is blocked, and the root raw response contained only 88 bytes. F20 medium: Canonical/indexing signals are incomplete |
be-discoverablestructured-and-shareable-metadata | not-applicable | high | A generic search-results listing is not an article, product, event, or other rich entity requiring JSON-LD; Open Graph metadata is present. |
be-private-and-securesecure-transport-and-headers | issues | high | HTTPS and CSP are present, but HSTS and nosniff are missing; three cookies were not Secure and every observed cookie used SameSite=None. F21 high: Transport defenses and cookie flags are incomplete |
be-private-and-securedata-minimisation-and-third-parties | issues | high | Tracker capture saw 15 third-party origins including Facebook; the homepage HAR attributed 117/119 requests and 5.67 MB to origins outside the exact host. F22 medium: The page has a broad third-party and tracking footprint |
be-private-and-securein-context-permissions-and-modern-auth | pass | medium | No geolocation/notification prompt appeared on load; authentication was not part of the covered search journey. |
be-private-and-securedefensive-browser-policies | issues | high | Headers evidence shows no HSTS, nosniff, or Permissions-Policy; frame protection and Referrer-Policy are present. F23 high: Several browser-enforced defenses are absent |
be-resilientprogressive-enhancement | pass | medium | Discoverability captured 98% raw/rendered word coverage for results, with title and description present without JavaScript. |
be-resilientresilient-runtime-behaviour | pass | medium | Search, horizontal categories, skeletons and result modules rendered at 360 px without clipping or horizontal page overflow. |
be-resilientoffline-and-installable | not-applicable | high | Search is intrinsically network-dependent; an offline/installable experience is not required for this public search surface. |
be-resilientnetwork-and-http-failure-states | pass | medium | The invalid route renders an explanatory Korean error with working back/home recovery choices rather than a blank shell. |
be-internationalisedlang-dir-and-logical-properties | pass | medium | The Korean page declares `lang="ko"`; mixed Korean/Latin query text renders in the expected left-to-right flow. |
be-internationalisedlocale-aware-data | not-applicable | high | The covered Korean search pages displayed no user-facing currencies, dates, durations, or locale-switchable numeric data requiring an Intl comparison. |
be-internationalisedtime-zone-correctness | not-applicable | high | The covered paths contained no scheduling, recurring events, or time-zone-sensitive user flow. |
be-trustworthyno-dark-patterns | pass | medium | Ads in the captured results are explicitly labeled and no consent wall, confirmshaming, forced continuity, or disguised primary action appeared. |
be-trustworthyhumane-error-handling | not-applicable | high | The primary form accepts free-text search and has no required-field validation or destructive submission to exercise. |
be-trustworthytrustworthy-input-assistance | not-applicable | high | No address, payment, sign-in, or sign-up fields occur in the covered public search flow; autocomplete=off on a search query is intentional. |
be-trustworthysafe-commercial-and-account-flows | not-applicable | high | Checkout, subscription, authentication, and account management are outside the anonymous search templates audited. |
be-sustainableoptimised-assets | issues | high | The image audit found no modern-format images, 25 images without srcset, 23 without dimensions, and six below-fold images not lazy-loaded. F24 high: Images are not delivered with an efficient responsive strategy |
be-sustainableno-wasteful-work | issues | high | The entry load made 119 requests/5.68 MB and included session, logging, AI, ad, and nudge requests plus multiple long tasks. F25 high: The page performs disproportionate network and background work |
be-sustainablethird-party-and-media-budget | issues | high | HAR measured 2.95 MB images, 1.62 MB scripts and 774 KB fonts; tracker evidence found 15 third-party origins. F26 high: Third-party, image, font, and script weight is disproportionate |
be-agent-readystructured-agent-capabilities | not-applicable | high | WebMCP is emerging and no declared agent-facing transactional capability exists; high raw-HTML result coverage already supports read-only agents. |
be-agent-readyon-device-inference | not-applicable | high | No user task in the covered search journey requires optional on-device inference; server search/AI briefing is the product behavior. |
be-memory-efficientno-leak-under-repeated-interaction | issues | medium | Across the repeated-focus probe JS heap rose from 55.2 MB to 89.9 MB and DOM elements rose 2,296→2,362; heap summaries differed by about 6 MB and 76k snapshot nodes. This is medium-confidence because background result modules may still be initializing. F27 medium: Repeated search-field interaction grows memory materially |
be-memory-efficientbounded-footprint | issues | medium | The baseline heap summary contained 1,406,829 snapshot nodes and 82.6 MB self size, including 95,861 closures and 11,549 FiberNode objects. F28 medium: The result page has a large initial heap footprint |
be-memory-efficientno-detached-dom-or-unbounded-listeners | pass | low | No Detached* constructor appeared among the heap summary’s leading constructors; DOM growth after ten focus cycles was limited to 66 elements. Confidence is low because listener instrumentation is unavailable. |
Provenance
Canonical report: results/atomic/reports/0251-m_search_naver_com.json
Report SHA-256: 9fc9b5252452a4394b6c1a1a87ae0ef3d73ba7e2f68852d7d54836eb486bbd52
Local retained report: runs/2026-07-17T17-27-24-856Z/atomic-reports/m_search_naver_com/2026-07-19T11-28-59-641Z/report.json
Local evidence root: runs/2026-07-17T17-27-24-856Z/atomic-reports/m_search_naver_com/2026-07-19T11-28-59-641Z
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.