Manifest position 459 · CrUX rank bucket 1000
https://www.lefigaro.fr
Coverage complete
Atomic coverage complete across 58 checks and six representative paths/templates. Checkout/payment and authenticated account internals were not entered, but the public subscription choice surface was audited.
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 byte-identical to the default capture (124,352 bytes), and the DOM reports color-scheme: normal with a white body background. F01 medium: The site does not adapt to the user’s dark colour preference. |
respect-user-preferencesrespects-reduced-motion | issues | high | The reduced-motion probe confirmed the preference and still found four running skeleton/skeletonShimmer animations, each with a 1,600 ms duration. F02 low: Loading animations continue under reduced-motion. |
respect-user-preferencesrespects-contrast | pass | high | Forced-colors/prefers-contrast screenshot preserved readable text and visible controls. |
implement-natural-interactionsview-transitions | issues | high | The captured DOM/CSS contains no view-transition rule or declaration across a multi-page news experience. F03 low: Navigation swaps are abrupt rather than using progressive-enhanced View Transitions. |
implement-natural-interactionsscroll-driven-animations | not-applicable | high | No scroll-linked parallax, scrollytelling or reveal interaction was present on the representative templates, so there is no scroll-linked implementation to judge. |
implement-natural-interactionsphysical-gestures | pass | high | Captured CSS/DOM includes scroll-snap use and no custom pointermove gesture layer was observed. |
provide-guided-navigationscroll-state-aware-chrome | issues | high | The homepage is about 32,000 CSS px tall, but captured CSS/DOM contains no scroll-state query or progress indicator; navigation chrome remains static. F04 low: Long news pages lack a scroll-state progress or responsive chrome cue. |
provide-guided-navigationanchored-positioning | issues | high | DOM/CSS inspection found no anchor-name, position-anchor or position-try use despite menu and panel controls. F05 low: Menus and transient panels do not expose CSS anchor-positioning/fallback placement. |
provide-guided-navigationdirects-attention | pass | high | Section branding, article breadcrumb text and the current bottom-navigation state clearly identify location. |
maximize-content-reduce-noiseno-intrusive-interruptions | issues | high | Desktop and 360 px screenshots of home, section, article and search show a large modal covering the headline or task, with additional sticky subscription chrome on mobile. F06 high: A consent wall obscures the entire first viewport on every representative page. |
maximize-content-reduce-noisesemantic-dismissible-primitives | issues | high | The platform probe found an iframe titled “Consent window”; the document contains no dialog element, while the blocking surface is rendered outside the document’s semantic controls. F07 medium: The consent surface is an iframe-based custom modal rather than a document-native dismissible primitive. |
maximize-content-reduce-noisereduced-chrome | issues | high | At 360×800 the consent surface, red subscription banner and bottom navigation leave no usable article content in the first viewport. F08 medium: Mobile content is crowded by stacked consent, subscription and fixed navigation chrome. |
adapt-to-the-form-factorresponsive-no-horizontal-scroll | pass | high | All four 360 px layout probes report scrollWidth=clientWidth=360, viewport metadata and zero horizontal overflow. |
adapt-to-the-form-factorcomponent-level-responsiveness | issues | high | The DOM probe reports container-type: normal and captured CSS contains no @container rule, although cards are reused in many widths. F09 low: Responsive behaviour is viewport-led rather than component-led. |
adapt-to-the-form-factorinput-modality-aware | issues | high | The focus probe found many 22 px-high navigation links and only a 1 px dotted focus outline; Lighthouse also failed target-size. F10 medium: Some navigation targets are too short and focus treatment is weak. |
support-core-task-successclear-purpose-and-primary-action | issues | high | Across home, section, article and search mobile captures, the consent wall obscures the page heading and primary content/task. F11 high: The page purpose and next action are hidden on first visit. |
support-core-task-successprimary-flow-completion | issues | high | The search page contains a labelled question textarea and submit action, but its first viewport is fully covered by consent; a direct ?question=agriculture URL does not execute or prefill the query. F12 high: Consent blocks the representative search journey before the user can ask a question. |
support-core-task-successclear-system-state-and-recovery | pass | high | The exercised 404 explains the error and offers clear routes back to the homepage or search. |
be-fast-and-stablegood-core-web-vitals | issues | high | Lighthouse measured LCP 10.7 s, FCP 4.2 s, TTI 16.5 s and performance 0.63; the independent trace measured LCP 2.57 s under a lighter desktop-like run, confirming strong condition sensitivity. F13 high: Mobile load performance is poor. |
be-fast-and-stablevisual-stability | pass | high | CLS was 0 on home mobile/article, 0.00026 on section and 0.00886 on search; desktop home was 0.0196, all below the good threshold. |
be-fast-and-stableefficient-main-thread | issues | high | The trace recorded a 664.93 ms longest task and 651.11 ms total blocking time across three long tasks. F14 high: A very long main-thread task blocks responsiveness. |
be-fast-and-stableefficient-resource-delivery | issues | high | The HAR recorded 225 requests and 2.54 MB transferred, including 86 scripts/1.95 MB and 91 stylesheets; multiple parser-inserted high-priority scripts originate in the document head. F15 high: Critical delivery is dominated by scripts and a large dependency graph. |
be-fast-and-stabletrim-unused-and-duplicate-code | issues | high | Five Google tag-manager/gtag resources account for roughly 1.39 MB, while third-party requests account for 109 requests and 2.02 MB. F16 high: Third-party and duplicated tag-manager work dominates the JavaScript budget. |
be-inclusivenames-roles-labels | issues | high | axe found critical required-parent/required-children and invalid aria-controls failures, unnamed listboxes and three unnamed carousel-dot buttons. F17 high: Core carousel and bottom-navigation semantics are invalid. |
be-inclusivesufficient-contrast | issues | high | axe measured examples at 4.29:1, 2.84:1 and 1.6:1, including ad labels, game titles and the TV-grid label. F18 medium: Several text labels fail WCAG AA contrast. |
be-inclusivestructure-and-focus | issues | high | axe reported critical invalid menubar/menuitem parenting, list items under a non-list role and nested interactive descendants; focus styling is only a 1 px dotted outline. F19 high: ARIA structure undermines focus and reading order. |
be-inclusivelegible-text | pass | high | Representative screenshots show readable body copy, clear hierarchy and no narrow-viewport clipping. |
be-inclusivezoom-reflow-targets-and-media | pass | high | Viewport allows scaling (no user-scalable=no), and all tested 360 px templates reflow without horizontal overflow. |
follow-best-practicesno-console-errors | pass | high | Lighthouse “errors-in-console” passed with an empty error table. |
follow-best-practicessound-document-and-assets | issues | high | Lighthouse failed image-aspect-ratio; the image primitive found 9 images without dimensions, 63 without srcset and 9 legacy-format images among 100 inspected. F20 medium: Some image assets have invalid sizing metadata. |
follow-best-practicesbrowser-platform-hygiene | issues | high | Lighthouse failed notification-on-start and traced the call to async-helpers JavaScript, before a user action. F21 medium: The page requests notification permission on load. |
be-discoverabletitle-and-description | pass | high | Home, section and article expose descriptive titles and meta descriptions; the article title and description match visible content. |
be-discoverablecrawlable-and-mobile-friendly | pass | high | Links use real href values, viewport metadata is present, robots allows ordinary public pages, and representative layouts are mobile-friendly. |
be-discoverablecanonical-and-indexing-signals | pass | high | Home/section/article returned 200 with canonical URLs and indexable robots metadata; robots.txt declares multiple dedicated sitemaps. |
be-discoverablestructured-and-shareable-metadata | pass | high | Home has Organization/WebSite/WebPage JSON-LD and complete Open Graph title, description, URL and 1200×1200 image; article metadata matches its visible headline. |
be-private-and-securesecure-transport-and-headers | issues | high | Headers show CSP limited to upgrade-insecure-requests; fig_save_consent and Wysistat lack Secure, and all three observed cookies use SameSite=None. F22 high: Transport is HTTPS, but CSP and cookie protection are weak. |
be-private-and-securedata-minimisation-and-third-parties | issues | high | The tracker probe saw 21 non-main origins and known Google, DoubleClick and Taboola trackers; HAR recorded 109 third-party requests/2.02 MB while analytics payloads state isConsentGiven=false. F23 high: A large tracking footprint loads before advertising consent. |
be-private-and-securein-context-permissions-and-modern-auth | issues | high | Lighthouse identified a notification permission request during page load rather than following a user gesture. F24 medium: Notification permission is requested out of context. |
be-private-and-securedefensive-browser-policies | issues | high | The main response lacks X-Content-Type-Options, Referrer-Policy and Permissions-Policy; CSP contains only upgrade-insecure-requests, though HSTS and X-Frame-Options are present. F25 medium: Several browser-enforced defenses are absent or incomplete. |
be-resilientprogressive-enhancement | pass | high | Raw HTML contains 98% of home and 100% of section/article rendered content words, with title/H1/description broadly surviving without JavaScript. |
be-resilientresilient-runtime-behaviour | pass | high | Representative templates loaded stable server-rendered content, menus remained within the narrow viewport and no runtime console error was reported. |
be-resilientoffline-and-installable | not-applicable | high | This is a server-rendered news publication rather than an installable application; a service worker controls the page, but installability is not required for the audited core reading journey. |
be-resilientnetwork-and-http-failure-states | pass | high | A nonexistent route produced an explicit Page introuvable state with explanation and recovery actions instead of a blank shell. |
be-internationalisedlang-dir-and-logical-properties | issues | high | Representative French pages declare lang=fr, but the blocking consent heading/body and “Continue without accepting” link are English while action labels are partly French. F26 medium: The French experience presents a mixed-language consent flow. |
be-internationalisedlocale-aware-data | pass | high | Visible dates/durations are rendered in French (for example “Il y a 32 minutes”), and prices use French euro formatting on subscription offers. |
be-internationalisedtime-zone-correctness | pass | high | The audited content uses relative publication duration rather than ambiguous bare local timestamps; no scheduling or cross-zone event flow is present. |
be-trustworthyno-dark-patterns | issues | high | “ACCEPT ALL” and setup are large filled buttons, while “Refuse all” is plain text at the bottom and “Continue without accepting” is a small top link; the modal blocks content. F27 high: Consent choices use asymmetric visual emphasis. |
be-trustworthyhumane-error-handling | not-applicable | high | The representative search control has no invalid-input state and the audited subscription selector uses labelled radio choices; there was no error-producing form validation flow to exercise. |
be-trustworthytrustworthy-input-assistance | not-applicable | high | The audited search and subscription-choice surfaces do not collect address, payment or sign-in data; autofill assistance is not applicable before the external checkout/auth step. |
be-trustworthysafe-commercial-and-account-flows | pass | high | Subscription offers clearly show monthly prices, one-year versus no-commitment choices, renewal prices, “résiliable à tout moment” terms and customer-service contact. |
be-sustainableoptimised-assets | issues | high | Of 100 images, 63 lack srcset, 66 below-fold images were not marked lazy and 9 use legacy formats; 9 lack dimensions. F28 medium: Image delivery leaves avoidable responsive-format waste. |
be-sustainableno-wasteful-work | issues | high | HAR shows 109 third-party requests and 2.02 MB third-party transfer, while the trace reports 651 ms blocking time and tracker calls indicate consent=false. F29 high: Advertising and analytics perform substantial work before consent. |
be-sustainablethird-party-and-media-budget | issues | high | Third parties transfer 2.02 MB of the 2.54 MB load; tag manager, ad-tech and recommendation scripts are the largest offenders. F30 high: Third-party code is disproportionate to the first-view news content. |
be-agent-readystructured-agent-capabilities | not-applicable | high | The robots policy explicitly rejects broad AI crawling/training and no developer intent to expose agent actions was declared, so emerging agent-tool exposure is outside this run’s intended surface. |
be-agent-readyon-device-inference | not-applicable | high | No declared product intent calls for client-side inference; the publication’s search is a server product and built-in AI adoption is an optional emerging opportunity. |
be-memory-efficientno-leak-under-repeated-interaction | pass | high | After ten next/previous carousel cycles and garbage collection, heap self-size fell from 45.61 MB to 44.74 MB and node count fell from 810,232 to 775,261. |
be-memory-efficientbounded-footprint | pass | high | The baseline retained self-size was 45.61 MB with 4,416 DOM elements; large but bounded for the media-heavy page and it did not grow after interaction. |
be-memory-efficientno-detached-dom-or-unbounded-listeners | pass | high | After repeated carousel interaction, native/object/closure populations and total nodes all decreased; no Detached constructor appeared among top retained constructors. |
Provenance
Canonical report: results/atomic/reports/0459-www_lefigaro_fr.json
Report SHA-256: 260c9b9e3d5e151c754648c5f8595579d0ec1ff0b0cf667e84dc8aa2d1336c50
Local retained report: runs/2026-07-17T17-27-24-856Z/atomic-reports/www_lefigaro_fr/2026-07-20T23-15-45-275Z/report.json
Local evidence root: runs/2026-07-17T17-27-24-856Z/atomic-reports/www_lefigaro_fr/2026-07-20T23-15-45-275Z
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.