On July 21, 2026, Claude stopped working for about an hour. The disruption started at 15:35 UTC and was resolved by 17:07 UTC. Anthropic has since marked it resolved. If you had workflows or an app running on Claude during that window, they went quiet with no warning.
Who this kind of outage hits
If you build on Claude, you probably are not watching its status page while you work. Most operators find out something broke the same way: a customer writes in, confused or annoyed, and only then does the scramble begin. By that point the disruption has already been running for a while, the customer has already formed an impression, and you are catching up instead of getting ahead of it. That gap, between when the tool stopped and when you heard about it, is where the damage happens.
Why it is especially rough without a technical background
When a tool like Claude goes quiet, there is no error message waiting for you. No log file, no alert, no red screen. Your automation just stops producing results. A workflow that usually runs every few minutes sits idle. A customer-facing feature returns nothing, or returns something stale. The first sign is often a confused message from someone who paid you. At that point you are not just fixing a problem, you are also managing a relationship.
Timeline
- 15:35 UTC - Claude stopped working. Major outage begins.
- ~16:35 UTC - About one hour into the disruption.
- 17:07 UTC - Claude recovered. Anthropic marked the issue resolved.
Total duration: approximately 1 hour and 32 minutes.
How a watcher catches this before your users do
Anthropic publishes a public status page. NoCrash reads that page every minute. The moment it flips from working to having trouble, NoCrash sends you a plain-language message, in words you can act on, without you having to go check anything yourself. You find out within a minute of Anthropic’s own report, not an hour later when a customer writes in.
NoCrash also watches the things you ship. If you have n8n workflows, it watches those too. If you have an app, you can give it a URL or drop in a small JS snippet and it will watch that as well. So if the problem is on your side rather than Claude’s, that surfaces too, in the same place.
To be clear about what this is: NoCrash reads public status and watches the specific things you list. It does not find outages before the tool’s own status page does. It just makes sure you are not the last to know.
For the authoritative account of this outage, see Anthropic’s own status page: https://stspg.io/74sqh6d7f9vq