Manifest position 393 · CrUX rank bucket 1000

https://www.temu.com

Coverage complete

Coverage-complete audit. Direct deep links repeatedly redirected to sign-in, so search/cart/error-path behavior is reported as observed rather than silently omitted.

Attempts

1 / 3

Judged checks

58 / 58

Blocked

Not run

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-preferences `respects-color-scheme`	issues	high	The dark-mode screenshot remains white/grey, and the CSS probe found no prefers-color-scheme rule. F01 medium: The interface does not honor the user’s dark color-scheme preference.
respect-user-preferences `respects-reduced-motion`	pass	high	Reduced-motion emulation matched and the active-animation probe returned an empty list.
respect-user-preferences `respects-contrast`	pass	high	Forced-colors screenshot retained legible dialog text, links and all three consent actions.
implement-natural-interactions `view-transitions`	pass	high	Rendered homepage CSS declares @view-transition { navigation:auto } and a 0.3s root transition group.
implement-natural-interactions `scroll-driven-animations`	not-applicable	high	No scroll-linked animation was present on the sampled pages, so implementation technique is not applicable.
implement-natural-interactions `physical-gestures`	pass	high	The product feed uses native document scrolling and CSS any-hover/none fallbacks; no page-level pointermove gesture override was observed.
provide-guided-navigation `scroll-state-aware-chrome`	issues	high	The long commerce feed has persistent multi-row chrome, but rendered CSS contains no scroll-state container query or scroll-timeline rule. F18 medium: Navigation chrome is not declaratively aware of scroll state.
provide-guided-navigation `anchored-positioning`	not-applicable	high	No tooltip, transient menu or anchored popover was present in the sampled states.
provide-guided-navigation `directs-attention`	pass	high	The visible product feed has labelled category/search/account/cart navigation and CSS cross-document view transitions to preserve route continuity.
maximize-content-reduce-noise `no-intrusive-interruptions`	issues	high	Desktop, mobile, search, cart and brand-assets screenshots all show a large modal covering the page before interaction; on mobile it is clipped. F02 high: A cookie wall obscures the primary content on every sampled archetype.
maximize-content-reduce-noise `semantic-dismissible-primitives`	issues	high	Desktop, mobile, search, cart and brand-assets screenshots all show a large modal covering the page before interaction; on mobile it is clipped. F02 high: A cookie wall obscures the primary content on every sampled archetype.
maximize-content-reduce-noise `reduced-chrome`	issues	high	Desktop, mobile, search, cart and brand-assets screenshots all show a large modal covering the page before interaction; on mobile it is clipped. F02 high: A cookie wall obscures the primary content on every sampled archetype.
adapt-to-the-form-factor `responsive-no-horizontal-scroll`	issues	high	The layout primitive measured scrollWidth 769px against visualViewportWidth 360px, a 409px overflow; the cookie actions are clipped in the mobile screenshot. F03 high: The homepage overflows horizontally on a 360px viewport.
adapt-to-the-form-factor `component-level-responsiveness`	issues	high	The DOM/CSS probe found no @container rules while the captured CSS contains many max-width viewport media queries. F17 low: Components rely on viewport breakpoints rather than container-level adaptation.
adapt-to-the-form-factor `input-modality-aware`	issues	high	The active focus probe reported outline:none and no box shadow for input, Continue, role=button controls and links; 34 of the first 50 controls were below 44px in one dimension. F04 high: Keyboard focus is not visibly indicated and many controls are undersized.
support-core-task-success `clear-purpose-and-primary-action`	pass	high	The mobile first viewport clearly presents Temu search, shopping categories, product cards, prices, ratings and add-to-cart controls.
support-core-task-success `primary-flow-completion`	issues	high	Direct search, cart and unknown-route requests all redirected to a generic sign-in page; the originally requested state was unavailable and the consent wall blocked interaction. F05 high: Representative shopping routes do not preserve a predictable browse-to-buy path.
support-core-task-success `clear-system-state-and-recovery`	issues	high	Direct search, cart and unknown-route requests all redirected to a generic sign-in page; the originally requested state was unavailable and the consent wall blocked interaction. F05 high: Representative shopping routes do not preserve a predictable browse-to-buy path.
be-fast-and-stable `good-core-web-vitals`	issues	high	Lighthouse measured 14.8s LCP, 7.0s FCP and 500ms TBT (performance 46); the raw-CDP trace independently found five long tasks and 200ms TBT. F06 high: The measured load is slow and main-thread heavy.
be-fast-and-stable `visual-stability`	pass	high	The layout observer measured CLS 0 with no recorded shifts during the capture window.
be-fast-and-stable `efficient-main-thread`	issues	high	Lighthouse measured 14.8s LCP, 7.0s FCP and 500ms TBT (performance 46); the raw-CDP trace independently found five long tasks and 200ms TBT. F06 high: The measured load is slow and main-thread heavy.
be-fast-and-stable `efficient-resource-delivery`	issues	high	HAR recorded 214 requests and 3.32MB transferred, including 2.68MB of script, four parser-inserted high-priority stylesheets and several 0/403/424/429 responses. F07 high: The initial load ships a large script-heavy dependency graph.
be-fast-and-stable `trim-unused-and-duplicate-code`	issues	high	HAR recorded 214 requests and 3.32MB transferred, including 2.68MB of script, four parser-inserted high-priority stylesheets and several 0/403/424/429 responses. F07 high: The initial load ships a large script-heavy dependency graph.
be-inclusive `names-roles-labels`	issues	high	Lighthouse found unnamed dialogs, prohibited ARIA, missing required role parents/children, label/name mismatch, invalid list structure and insufficient contrast; accessibility scored 78. The image primitive found all 37 images without width/height, 17 without alt and 35 below-fold images without lazy loading; Lighthouse also flagged wrong aspect ratios and low-resolution delivery. F08 high: The rendered interface has multiple accessibility-tree and contrast failures. F09 medium: Images lack intrinsic dimensions and meaningful alternatives.
be-inclusive `sufficient-contrast`	issues	high	Lighthouse found unnamed dialogs, prohibited ARIA, missing required role parents/children, label/name mismatch, invalid list structure and insufficient contrast; accessibility scored 78. F08 high: The rendered interface has multiple accessibility-tree and contrast failures.
be-inclusive `structure-and-focus`	issues	high	The active focus probe reported outline:none and no box shadow for input, Continue, role=button controls and links; 34 of the first 50 controls were below 44px in one dimension. Lighthouse found unnamed dialogs, prohibited ARIA, missing required role parents/children, label/name mismatch, invalid list structure and insufficient contrast; accessibility scored 78. F04 high: Keyboard focus is not visibly indicated and many controls are undersized. F08 high: The rendered interface has multiple accessibility-tree and contrast failures.
be-inclusive `legible-text`	pass	high	Product names, prices, ratings and consent copy remain readable in the sampled desktop/mobile captures without text clipping inside content cards.
be-inclusive `zoom-reflow-targets-and-media`	issues	high	The layout primitive measured scrollWidth 769px against visualViewportWidth 360px, a 409px overflow; the cookie actions are clipped in the mobile screenshot. The active focus probe reported outline:none and no box shadow for input, Continue, role=button controls and links; 34 of the first 50 controls were below 44px in one dimension. F03 high: The homepage overflows horizontally on a 360px viewport. F04 high: Keyboard focus is not visibly indicated and many controls are undersized.
follow-best-practices `no-console-errors`	issues	high	Lighthouse reported console errors, one deprecated API, a BFCache failure, forced reflow and a failed network dependency-tree insight; best practices scored 69. F10 medium: The page emits browser errors and blocks platform optimizations.
follow-best-practices `sound-document-and-assets`	issues	high	The image primitive found all 37 images without width/height, 17 without alt and 35 below-fold images without lazy loading; Lighthouse also flagged wrong aspect ratios and low-resolution delivery. F09 medium: Images lack intrinsic dimensions and meaningful alternatives.
follow-best-practices `browser-platform-hygiene`	issues	high	Lighthouse reported console errors, one deprecated API, a BFCache failure, forced reflow and a failed network dependency-tree insight; best practices scored 69. F10 medium: The page emits browser errors and blocks platform optimizations.
be-discoverable `title-and-description`	issues	high	Discoverability measured 0% raw/rendered word coverage and classified the response as a JS shell; crawler screenshot is blank. robots.txt is a JavaScript challenge, and Lighthouse flagged invalid robots and canonical signals. F11 high: Public content is effectively invisible to non-JavaScript crawlers.
be-discoverable `crawlable-and-mobile-friendly`	issues	high	Discoverability measured 0% raw/rendered word coverage and classified the response as a JS shell; crawler screenshot is blank. robots.txt is a JavaScript challenge, and Lighthouse flagged invalid robots and canonical signals. F11 high: Public content is effectively invisible to non-JavaScript crawlers.
be-discoverable `canonical-and-indexing-signals`	issues	high	Discoverability measured 0% raw/rendered word coverage and classified the response as a JS shell; crawler screenshot is blank. robots.txt is a JavaScript challenge, and Lighthouse flagged invalid robots and canonical signals. F11 high: Public content is effectively invisible to non-JavaScript crawlers.
be-discoverable `structured-and-shareable-metadata`	issues	high	Discoverability measured 0% raw/rendered word coverage and classified the response as a JS shell; crawler screenshot is blank. robots.txt is a JavaScript challenge, and Lighthouse flagged invalid robots and canonical signals. F11 high: Public content is effectively invisible to non-JavaScript crawlers.
be-private-and-secure `secure-transport-and-headers`	issues	high	HTTPS and HSTS are present, but CSP, nosniff, frame protection, Referrer-Policy and Permissions-Policy are absent. Cookie __tclr lacks Secure and every observed cookie uses SameSite=None. F12 high: Important browser-enforced security policies are missing.
be-private-and-secure `data-minimisation-and-third-parties`	issues	high	HAR attributed 145 requests and 2.95MB to third-party origins; the tracker probe found nine third-party origins including Facebook, plus 49 pftk and 22 thtk requests. F13 high: Third-party and tracking activity dominates the initial request surface.
be-private-and-secure `in-context-permissions-and-modern-auth`	issues	high	The sign-in surface offers email/phone continuation; the form/source probe found no passkey/WebAuthn registration or authentication affordance despite browser support. F19 medium: The sampled account entry does not expose phishing-resistant authentication.
be-private-and-secure `defensive-browser-policies`	issues	high	HTTPS and HSTS are present, but CSP, nosniff, frame protection, Referrer-Policy and Permissions-Policy are absent. Cookie __tclr lacks Secure and every observed cookie uses SameSite=None. F12 high: Important browser-enforced security policies are missing.
be-resilient `progressive-enhancement`	issues	high	Raw HTML contains no content, the crawler view is blank, and both robots.txt and app.webmanifest return challenge HTML. The unknown route redirects to login instead of a useful error state. F14 high: Core content and machine-readable control files depend on JavaScript challenges.
be-resilient `resilient-runtime-behaviour`	issues	high	Desktop, mobile, search, cart and brand-assets screenshots all show a large modal covering the page before interaction; on mobile it is clipped. The layout primitive measured scrollWidth 769px against visualViewportWidth 360px, a 409px overflow; the cookie actions are clipped in the mobile screenshot. F02 high: A cookie wall obscures the primary content on every sampled archetype. F03 high: The homepage overflows horizontally on a 360px viewport.
be-resilient `offline-and-installable`	issues	high	Raw HTML contains no content, the crawler view is blank, and both robots.txt and app.webmanifest return challenge HTML. The unknown route redirects to login instead of a useful error state. F14 high: Core content and machine-readable control files depend on JavaScript challenges.
be-resilient `network-and-http-failure-states`	issues	high	Direct search, cart and unknown-route requests all redirected to a generic sign-in page; the originally requested state was unavailable and the consent wall blocked interaction. Raw HTML contains no content, the crawler view is blank, and both robots.txt and app.webmanifest return challenge HTML. The unknown route redirects to login instead of a useful error state. F05 high: Representative shopping routes do not preserve a predictable browse-to-buy path. F14 high: Core content and machine-readable control files depend on JavaScript challenges.
be-internationalised `lang-dir-and-logical-properties`	pass	high	Rendered HTML declares lang=en-GB; CSS probe found logical properties and locale-specific UK routes.
be-internationalised `locale-aware-data`	pass	high	The UK locale renders pound-denominated prices and en-GB content, and the page configuration exposes region/language/currency locale switching.
be-internationalised `time-zone-correctness`	pass	high	The page maintains an explicit timezone cookie/config signal and no ambiguous event time or recurring schedule was present in sampled surfaces.
be-trustworthy `no-dark-patterns`	pass	high	The consent dialog gives Reject All and Accept All equal-size, equal-emphasis controls plus Customise Cookies; no preselected paid continuity was observed.
be-trustworthy `humane-error-handling`	issues	high	The email/phone input reports required:false and an empty autocomplete token; the attempted empty submission remained blocked by the consent dialog, so actionable validation could not be reached. F15 medium: The sign-in form lacks visible native constraints and input assistance.
be-trustworthy `trustworthy-input-assistance`	issues	high	The email/phone input reports required:false and an empty autocomplete token; the attempted empty submission remained blocked by the consent dialog, so actionable validation could not be reached. F15 medium: The sign-in form lacks visible native constraints and input assistance.
be-trustworthy `safe-commercial-and-account-flows`	issues	high	Direct search, cart and unknown-route requests all redirected to a generic sign-in page; the originally requested state was unavailable and the consent wall blocked interaction. F05 high: Representative shopping routes do not preserve a predictable browse-to-buy path.
be-sustainable `optimised-assets`	issues	high	The image primitive found all 37 images without width/height, 17 without alt and 35 below-fold images without lazy loading; Lighthouse also flagged wrong aspect ratios and low-resolution delivery. F09 medium: Images lack intrinsic dimensions and meaningful alternatives.
be-sustainable `no-wasteful-work`	issues	high	Lighthouse measured 14.8s LCP, 7.0s FCP and 500ms TBT (performance 46); the raw-CDP trace independently found five long tasks and 200ms TBT. HAR recorded 214 requests and 3.32MB transferred, including 2.68MB of script, four parser-inserted high-priority stylesheets and several 0/403/424/429 responses. F06 high: The measured load is slow and main-thread heavy. F07 high: The initial load ships a large script-heavy dependency graph.
be-sustainable `third-party-and-media-budget`	issues	high	HAR recorded 214 requests and 3.32MB transferred, including 2.68MB of script, four parser-inserted high-priority stylesheets and several 0/403/424/429 responses. HAR attributed 145 requests and 2.95MB to third-party origins; the tracker probe found nine third-party origins including Facebook, plus 49 pftk and 22 thtk requests. F07 high: The initial load ships a large script-heavy dependency graph. F13 high: Third-party and tracking activity dominates the initial request surface.
be-agent-ready `structured-agent-capabilities`	issues	high	Lighthouse agentic-browsing scored 33, the raw response is an empty JS shell, and the probe found no structured tool surface or machine-readable form metadata beyond rendered controls. F20 medium: The commerce capabilities are not exposed in a structured agent-readable form.
be-agent-ready `on-device-inference`	not-applicable	high	No summarisation, generation or other inference use case was exposed on the sampled commerce and account surfaces.
be-memory-efficient `no-leak-under-repeated-interaction`	issues	low	The baseline heap summary had 433,162 nodes/25.64MB; after ten bottom-to-top scroll cycles it had 491,443 nodes/34.54MB, increases of 58,281 nodes and 8.89MB. Separate primitive sessions and route variability reduce confidence. F16 medium: Repeated scrolling correlates with substantial retained heap growth.
be-memory-efficient `bounded-footprint`	issues	low	The baseline heap summary had 433,162 nodes/25.64MB; after ten bottom-to-top scroll cycles it had 491,443 nodes/34.54MB, increases of 58,281 nodes and 8.89MB. Separate primitive sessions and route variability reduce confidence. F16 medium: Repeated scrolling correlates with substantial retained heap growth.
be-memory-efficient `no-detached-dom-or-unbounded-listeners`	pass	high	Neither readable heap summary listed Detached* constructors among retained populations; no direct detached-DOM signal was found.

Provenance

Canonical report: results/atomic/reports/0393-www_temu_com.json
Report SHA-256: 02cfc87fee631fbf53af2906e74ee241fc8e595e84b67e5907a521f3f08563b3
Local retained report: runs/2026-07-17T17-27-24-856Z/atomic-reports/www_temu_com/2026-07-20T11-21-22-947Z/report.json
Local evidence root: runs/2026-07-17T17-27-24-856Z/atomic-reports/www_temu_com/2026-07-20T11-21-22-947Z
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.