Manifest position 55 · CrUX rank bucket 1000

https://policies.google.com

Coverage complete

Atomic coverage complete across 58 checks and six representative paths.

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`	pass	high	Light and dark screenshots show complete surface and control retinting.
respect-user-preferences `respects-reduced-motion`	pass	high	Under prefers-reduced-motion: reduce, the active animation count was 0.
respect-user-preferences `respects-contrast`	pass	high	The high-contrast capture preserves legible text, controls, links, and iconography; Lighthouse accessibility scored 1.00.
implement-natural-interactions `view-transitions`	issues	high	The page probe found no view-transition CSS on a multi-document navigation surface. F01 low: Page-to-page navigation uses abrupt document swaps rather than view transitions.
implement-natural-interactions `scroll-driven-animations`	not-applicable	high	Active review of the representative paths found the stated outcome satisfied for Scroll-linked motion (parallax, scrollytelling, entry/exit reveals) uses declarative CSS scroll-driven animations (off main thread) instead of scroll event listeners.
implement-natural-interactions `physical-gestures`	not-applicable	high	Active review of the representative paths found the stated outcome satisfied for Gesture-driven interactions and entry/exit motion feel native (declarative overscroll/scroll-snap, physics-based easing, animating to intrinsic sizes, pull/swipe gestures) rather than fighting the platform with custom pointer handlers.
provide-guided-navigation `scroll-state-aware-chrome`	issues	medium	The privacy document measured 107,024 CSS px but the mobile chrome exposes no progress cue. F02 low: Very long policy pages do not expose visible reading progress in the mobile chrome.
provide-guided-navigation `anchored-positioning`	not-applicable	high	Active review of the representative paths found the stated outcome satisfied for Tooltips, popovers and menus use CSS anchor positioning (with fallback positions) so they stay attached and reposition correctly rather than being manually positioned.
provide-guided-navigation `directs-attention`	pass	high	Descriptive global links, prominent headings, and visible underlined policy actions establish clear navigation targets.
maximize-content-reduce-noise `no-intrusive-interruptions`	pass	high	Load screenshots across five templates show content immediately with no interstitial or obscuring consent wall.
maximize-content-reduce-noise `semantic-dismissible-primitives`	not-applicable	high	Active review of the representative paths found the stated outcome satisfied for Overlays and rich controls use the right primitive: popover (with declarative light-dismiss) for transient UI, dialog for modal flows, details for disclosure, native-but-branded selects and pickers, rather than ad-hoc divs.
maximize-content-reduce-noise `reduced-chrome`	pass	high	Screenshots show a compact single-row header and content-first policy layouts without decorative application framing.
adapt-to-the-form-factor `responsive-no-horizontal-scroll`	pass	high	At 360px the entry, privacy, and terms layouts reported 0px horizontal overflow and CLS 0.
adapt-to-the-form-factor `component-level-responsiveness`	issues	high	Mobile screenshots visibly truncate policy titles and the CSS probe found no @container usage. F03 medium: The mobile header clips or truncates the current-page title.
adapt-to-the-form-factor `input-modality-aware`	pass	high	Lighthouse accessibility and target-size audits passed; controls retain native semantic button/link behavior.
support-core-task-success `clear-purpose-and-primary-action`	pass	high	The first viewport names Privacy Policy and Terms of Service and gives descriptive “Read…” links.
support-core-task-success `primary-flow-completion`	pass	high	Navigation from the overview to Privacy, Terms, Technologies, and FAQ resolves to complete, readable documents without an account wall.
support-core-task-success `clear-system-state-and-recovery`	issues	high	The 404 response has no link or normal navigation for recovery. F04 medium: The 404 state explains the failure but offers no recovery action.
be-fast-and-stable `good-core-web-vitals`	issues	high	Lighthouse mobile LCP was 3,906 ms, above the 2.5 s good threshold. F05 medium: Mobile lab LCP is above the good Core Web Vitals threshold.
be-fast-and-stable `visual-stability`	pass	high	All measured 360px templates reported CLS 0 with no observed layout-shift entries.
be-fast-and-stable `efficient-main-thread`	pass	high	The trace found 0 long tasks and 0 ms total blocking time; Lighthouse TBT was 7 ms.
be-fast-and-stable `efficient-resource-delivery`	issues	high	HAR found a high-priority parser script and 108,729 font bytes in the critical load. F06 medium: A parser-inserted high-priority script and font chain compete with initial content.
be-fast-and-stable `trim-unused-and-duplicate-code`	issues	high	Lighthouse estimated 248,941 unused JS bytes and 36,629 unused CSS bytes. F07 medium: The policy hub ships substantial JavaScript that is unused during initial load.
be-inclusive `names-roles-labels`	pass	high	Lighthouse accessibility scored 1.00; image-alt and link-name audits passed.
be-inclusive `sufficient-contrast`	pass	high	Lighthouse accessibility scored 1.00 and light, dark, and contrast captures remain legible.
be-inclusive `structure-and-focus`	pass	high	Lighthouse heading-order, names, and accessibility audits passed; the DOM contains a skip-to-main-content control.
be-inclusive `legible-text`	pass	high	Mobile captures show comfortable line lengths, spacing, and unclipped article body text in English and Arabic.
be-inclusive `zoom-reflow-targets-and-media`	pass	high	Viewport allows scaling to 5, layouts reflow at 360px with no overflow, and Lighthouse target-size passed.
follow-best-practices `no-console-errors`	pass	high	Lighthouse best-practices scored 1.00 and did not report console-error or runtime-exception failures.
follow-best-practices `sound-document-and-assets`	pass	high	DOM reports HTML doctype, UTF-8, viewport meta; Lighthouse best-practices scored 1.00 and image aspect audits passed.
follow-best-practices `browser-platform-hygiene`	issues	high	Lighthouse found Cache-Control no-store prevents BFCache. F08 low: Cache-Control behavior prevents back/forward cache restoration.
be-discoverable `title-and-description`	issues	high	DOM and Lighthouse found no meta description. F09 medium: The policy hub has no meta description.
be-discoverable `crawlable-and-mobile-friendly`	pass	high	All 34 observed anchors had hrefs, none used generic link text, viewport meta is present, and raw HTML contains full content.
be-discoverable `canonical-and-indexing-signals`	issues	high	Lighthouse rejected the locale-query canonical; robots.txt and sitemap.xml returned 404. F10 low: The homepage canonical signal is not accepted as valid by Lighthouse.
be-discoverable `structured-and-shareable-metadata`	issues	high	Probe found zero Open Graph tags and zero JSON-LD blocks. F11 low: The public policy hub has no Open Graph or structured entity metadata.
be-private-and-secure `secure-transport-and-headers`	pass	high	HTTPS, CSP, nosniff, secure cookies and SameSite attributes were observed; no mixed content or sensitive client secret was found.
be-private-and-secure `data-minimisation-and-third-parties`	issues	high	Tracker capture found seven third-party origins and active GTM/Analytics. F12 medium: A privacy-policy hub loads analytics and seven third-party origins on first view.
be-private-and-secure `in-context-permissions-and-modern-auth`	not-applicable	high	Active review of the representative paths found the stated outcome satisfied for Permission prompts are requested in context (not on load), and authentication uses modern, phishing-resistant methods (passkeys / WebAuthn) where relevant rather than passwords alone.
be-private-and-secure `defensive-browser-policies`	issues	high	Headers lacked HSTS, Referrer-Policy, and framing restrictions. F13 medium: Several defensive response policies are absent or permissive.
be-resilient `progressive-enhancement`	pass	high	Discoverability captures found 100% rendered-content coverage in raw HTML on overview and Privacy Policy, with no JS shell.
be-resilient `resilient-runtime-behaviour`	pass	high	Five templates loaded without clipped content or runtime failures at 360px; no overlay or async control was present to destabilize.
be-resilient `offline-and-installable`	not-applicable	high	Active review of the representative paths found the stated outcome satisfied for Where the site is an app, it is installable (web app manifest) and offers an offline fallback and works on flaky networks. (Contextual: a brochure or intrinsically-online site may reasonably not need this.)
be-resilient `network-and-http-failure-states`	pass	high	The invalid URL returns a truthful HTTP 404 and explicit “requested URL was not found” state.
be-internationalised `lang-dir-and-logical-properties`	pass	high	English emits lang=en-US/dir=ltr and Arabic emits lang=ar/dir=rtl with correctly mirrored navigation and readable RTL layout.
be-internationalised `locale-aware-data`	pass	high	The Arabic Terms route translates and formats “17 March 2025” in Arabic reading order instead of retaining English copy.
be-internationalised `time-zone-correctness`	not-applicable	high	Active review of the representative paths found the stated outcome satisfied for Time handling survives time zones and DST: events coordinate across zones, partial time concepts are modelled, and stored times are unambiguous.
be-trustworthy `no-dark-patterns`	pass	high	No consent wall, upsell, disguised ad, forced continuity, or deceptive default appeared on any representative policy path.
be-trustworthy `humane-error-handling`	not-applicable	high	Active review of the representative paths found the stated outcome satisfied for Forms prevent and recover from mistakes humanely: validate after interaction (not prematurely), give clear required-field feedback, announce errors accessibly, and signal invalid fields visibly rather than blaming the user.
be-trustworthy `trustworthy-input-assistance`	not-applicable	high	Active review of the representative paths found the stated outcome satisfied for Input is assisted, not obstructed: correct autocomplete tokens so address, payment, sign-in and sign-up fields autofill, and inputs are highlighted/sized to help the user rather than trip them up.
be-trustworthy `safe-commercial-and-account-flows`	not-applicable	high	Active review of the representative paths found the stated outcome satisfied for Checkout, subscription, consent, authentication and account-management flows are clear, reversible, and proportionate: pricing and commitments are visible, cancellation is findable, sensitive actions re-authenticate when appropriate, and users are not tricked into continuity.
be-sustainable `optimised-assets`	issues	high	Images probe found six images without intrinsic dimensions and two below-fold images not lazy-loaded. F14 low: Small decorative images omit intrinsic dimensions and below-fold lazy loading.
be-sustainable `no-wasteful-work`	issues	high	Measured unused JS/CSS and eager analytics represent avoidable transfer/work. F07 medium: The policy hub ships substantial JavaScript that is unused during initial load.
be-sustainable `third-party-and-media-budget`	issues	high	Seven third-party origins and a 159,899-byte tag-manager script are disproportionate for static policy content. F12 medium: A privacy-policy hub loads analytics and seven third-party origins on first view.
be-agent-ready `structured-agent-capabilities`	not-applicable	high	Active review of the representative paths found the stated outcome satisfied for Where it makes sense, the site exposes structured, safe capabilities to agents via WebMCP tools, agentic forms, and agentic JavaScript tools rather than leaving agents to scrape and guess.
be-agent-ready `on-device-inference`	not-applicable	high	Active review of the representative paths found the stated outcome satisfied for On-device inference (built-in language model, summariser) is used appropriately where it improves the experience, rather than shipping every task to a server.
be-memory-efficient `no-leak-under-repeated-interaction`	not-applicable	high	Active review of the representative paths found the stated outcome satisfied for Repeating a representative interaction (open/close a modal, navigate a route and back, infinite-scroll a list) about 10 times does not grow retained heap without bound; what is allocated during the interaction is released when it ends.
be-memory-efficient `bounded-footprint`	pass	high	The baseline heap was 252,506 nodes and 11,749,279 self-size bytes, proportionate for this content MPA.
be-memory-efficient `no-detached-dom-or-unbounded-listeners`	pass	high	The heap constructor summary contained no Detached*, EventListener, or Timeout constructor among reported size leaders.

Provenance

Canonical report: results/atomic/reports/0055-policies_google_com.json
Report SHA-256: 15763ef62a853132553615d3d57611bca80a9c34b5896dad5493cfba82970a22
Local retained report: runs/2026-07-17T17-27-24-856Z/atomic-reports/policies_google_com/2026-07-18T02-51-28-804Z/report.json
Local evidence root: runs/2026-07-17T17-27-24-856Z/atomic-reports/policies_google_com/2026-07-18T02-51-28-804Z
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.