Observed
A specific response arrived.
The evidence applies only to this request. It does not establish availability, speed or the outcome of the next call.
Clarity Status
Operational status starts with evidence.
This page separates observed delivery, existing configuration and preview. It shows availability only when independent measurement points provide real data.
Status preview
No green light without independent measurement.
Clarity does not currently publish a combined ‘all systems operational’ state. External probes, measurement history, alerting and a public incident channel are not yet connected.
Status language
Three words. Three clearly different statements.
None of these states automatically means ‘healthy’. The label describes only the kind of evidence available today.
Configured
Prerequisites are in place.
Required server configuration or approval gates exist. This does not measure whether the service is reachable and works correctly.
Preview
The surface describes a target.
The service or its public operational evidence has not yet been released. A preview is neither an outage nor an operational promise.
Components
What this page establishes. And what it explicitly does not.
Clarity website
Available evidence
This status page was delivered for the current request.
Not established
No external uptime evidence, no history and no statement about other pages or the next request.
Observed
Clarity Account
Available evidence
A validly formed backend endpoint and public client access are configured on this server.
Not established
No public synthetic login, independent health check or published latency measurement.
Configured
Clarity support · human
Available evidence
Backend, minors and legal gates for new human support cases are configured.
Not established
No external case test, public queue health check or response-time promise.
Configured
Clarity support · AI-first
Available evidence
The separate AI, privacy, provider, minors and backend gate is also configured.
Not established
Configuration is not external evidence for drafting, the five-minute hold, override or automatic delivery.
Configured
Developer APIs
Available evidence
Documented target architectures and clearly marked Private Preview surfaces exist.
Not established
No public API, approved keys, webhooks, quotas or availability promise.
Preview
Target model · not yet live
From a request to a dependable statement.
A status becomes publicly dependable only when the same traceable chain is defined, operated and reviewed for every component.
Probe
An independent external measurement calls a defined endpoint. Synthetic accounts must not contain real user data.
Signal
Time, location, result and duration are recorded for each component, rather than inferred from a single browser visit.
Decision
Only published success criteria, maintenance rules and alert thresholds may turn signals into an operational state.
Communication
A confirmed state names the affected component, impact, start time and the time of the next dependable update.
Incident lifecycle · target process
An incident needs more than a red symbol.
This process describes the intended publication standard. A public incident history begins with verified events, not demo entries.
Detect
A connected monitor or verifiable report provides the first signal.
Confirm
The signal is reproduced, narrowed down and assigned to a responsible component.
Contain
Impact and reach are limited; safe workarounds are named only when they have been verified.
Resolve
The cause is corrected and recovery is verified through the same measurement route.
Review
The timeline, cause, impact and concrete prevention are documented after review.
Monitoring boundaries
What is not measured here today.
- No independent multi-region or mobile-network measurement point.
- No public history for uptime, response time or error rate.
- No synthetic end-to-end test for sign-in, Account or support cases.
- No published SLO, maintenance or alerting definition.
- No statement about internal queue capacity or individual handling times.
Status updates
No status subscription yet.
Email, RSS or webhook notifications will be offered only when the same verified incident channel also supplies this page. A placeholder sign-up would not deliver dependable updates.
What you can do today
Report a problem. Keep the security route separate.
A specific product or account problem belongs in support. Possible vulnerabilities follow the separate confidential security route.