On the evening of July 28, 2026, Bolt.new had a significant disruption. It started at 21:35 UTC and lasted about 33 minutes, recovering at 22:08 UTC. Bolt.new has since reported it resolved.
Who this kind of outage hits
If you build on Bolt.new, you probably found out the same way most people do: a customer messaged you, or you went to check on something and noticed it was broken. The tool itself rarely sends you a note saying “hey, we’re having trouble right now.” You are left piecing it together after the fact, often hours after the problem started. That gap, between when the tool stopped working and when you found out, is where the damage happens.
Why it is especially rough without a technical background
When you are not an engineer, there is no log to open, no error message to decode. The work just stops moving. A build that should have finished sits there. A customer tries something and gets nothing. The first real signal is usually an unhappy message from someone who paid you. By then you are already behind, already apologizing, already explaining something you only just learned yourself. That is a hard position to be in, and it is the default for most solo operators and small teams.
What the timeline looked like
- 21:35 UTC – Bolt.new started experiencing a bigger problem.
- 22:08 UTC – Bolt.new reported the issue resolved.
- Total duration – about 33 minutes.
How a watcher catches this before your users do
NoCrash reads Bolt.new’s public status page once every minute. The moment that page flips from working to having trouble, NoCrash sends you a plain-language note, in words you can act on, without you having to go check anything yourself. That turns “my customer told me an hour later” into “I got a calm heads-up within a minute of Bolt.new’s own report.” It sits next to everything else you build on, so you are not juggling five different status pages.
NoCrash also watches the things you ship. It can watch your n8n workflows through an API token, and your app through a URL you give it or a small JS snippet you drop in. So if something goes quiet on your own side, that surfaces too, separately from what the underlying tool is reporting.
To be clear about what it does not do: NoCrash reads the tool’s own public status page. It does not find the outage before the tool reports it. It catches the report within a minute and gets it to you in plain English, which is usually well before a customer notices and reaches out.
For the authoritative account of this outage, see Bolt.new’s own status page at https://status.bolt.new/proxy/status.bolt.new.