On June 23, 2026, Bolt.new reported a bigger problem starting at 16:37 UTC. It lasted about three hours and was resolved at 20:16 UTC. Bolt.new has since confirmed the disruption is over.
Who this kind of outage hits
If you build or ship something on Bolt.new, you probably found out the wrong way. Not from a log, not from an alert. From a customer saying something felt broken, or from noticing yourself that a project you were working on had quietly stopped responding. That gap, between when the tool went quiet and when you heard about it, is where the damage happens. A paying user files a support ticket. A client loses confidence. You spend the first twenty minutes just trying to figure out whether the problem is on your side or theirs.
Why this is especially rough without a technical background
When something stops working inside a tool you build on, there is usually nothing on your screen that explains it. No red banner, no error message you can act on. The work just stops moving. If you are not an engineer, you have no logs to dig through and no way to tell whether the problem is a five-minute blip or a three-hour outage. The first real signal is often an unhappy customer, and by then you are already behind.
What the timeline looked like
- 16:37 UTC - Bolt.new reported a bigger problem.
- About 3 hours - the disruption continued.
- 20:16 UTC - Bolt.new confirmed the issue was resolved.
Bolt.new has not published a detailed cause as of this writing. The official source below is the place to check for any further updates.
How a watcher catches this before your users do
Bolt.new has a public status page. It updates when something goes wrong. The problem is that most people are not watching it, and even if they bookmarked it once, they are not refreshing it every minute while they work.
NoCrash reads that status page every minute. The moment Bolt.new’s own page flips from working to having trouble, NoCrash sends you a plain-language heads-up, in words you can act on, sitting next to everything else you build on. You find out within a minute of Bolt.new’s own report, not an hour later when a customer writes in.
It also watches the things you ship. If you have n8n workflows, NoCrash watches those too. If you have an app or a URL you want watched, you can give it that. So if something goes quiet on your own side, that surfaces as well. What it does not do is claim to know about an outage before the tool’s own status page says so. It reads what Bolt.new publishes, translates it, and gets it to you fast.
The result is that “my customer told me something was broken” becomes “I got a calm heads-up two hours before my customer noticed.”
For the authoritative account of this outage, see the official Bolt.new status page: https://status.bolt.new/proxy/status.bolt.new