First probe pending
We're running your check now. Reload in a few seconds to see live data.
Publisher status · statuspage
Anthropic / Claude status
Live incidents and maintenance pulled from the publisher's official status page.
PENDING
Checked 1s ago
single-source feed
every 5 min · publisher feed
direct probe, not crowd-reported
Publisher status
- Indicator
- error
- Description
- Bad JSON: Unexpected token '<', "<html><bod"... is not valid JSON
- Source
- https://status.anthropic.com
- Adapter
- statuspage
What this means
Publisher status data not yet available.
A poll is in progress or the upstream feed returned an error. Refresh in a few seconds.
What to try next
Browse
Explore the global status page while you wait.
Official status page
status.anthropic.com Anthropic / Claude's official status pageFAQ
Why does this page show Anthropic / Claude as up while my game won't connect?
Publisher status pages show the publisher's view, not yours. Anthropic / Claude aggregates many internal signals; user-visible problems (matchmaker queues, regional CDN issues, account auth) sometimes don't trip the public indicator. Check the active incidents list below for narrower scope.
How is this different from a direct probe?
Websites and game servers get hit with a live network request. Publishers don't expose a single check endpoint, so we poll their official status feed (Statuspage, Riot, etc.) every 5 minutes and surface their classification verbatim.
Where does the data come from?
Directly from Anthropic / Claude's own status feed — see https://status.anthropic.com. We don't aggregate Twitter or DownDetector signals.
Status badge
Embed a live status badge on your site, README, or status page. Updates automatically — same status-aware SVG the rest of isitdown.io uses.
<a href="https://isitdown.io/publisher/anthropic"><img src="https://isitdown.io/badge/publisher/anthropic.svg" alt="Anthropic / Claude status"></a>
Related
Methodology
- Probe
- We poll the publisher's official status page endpoint (Statuspage.io JSON, Riot's status API, etc.) and read their own incident feed.
- Counts as DOWN
- The publisher reports a major or critical indicator, or the adapter returns a protocol error.
- Counts as DEGRADED
- The publisher reports minor issues or maintenance. Active incidents are listed even when the overall indicator is green.
- Detail
- Publisher checks differ from direct probing: we inherit the publisher's own classification. If Atlassian says "all systems operational" while users can't connect, our check will show up — the signal is theirs, not ours.
- Cadence
- Every 5 minutes, in parallel across 4 monitoring regions (US East Virginia, US West Oregon, Europe London, Asia Singapore).
- Rate-limited targets
- If a host returns 429 or consistently drops connections from our IPs, we cap retries at 3 and report the last observed status — we do not flood the target to confirm the outage.
- Data source
- Direct probes from our monitoring infrastructure. We do not aggregate crowd reports, Twitter mentions, or DownDetector signals — every result on this page is a live network request.