Snagtrail catches errors in your app, groups the repeats into one issue, and points at the exact line and the commit that changed it.
Works with Node, Next.js, Remix, Bun and the browser. Alerts in Slack or email.
Three things happen the moment an error leaves your app. You only see the last one.
Add three lines and every crash, rejected promise and failed request is caught, along with the release and the browser it happened in.
See the installThe same crash hitting twelve customers is one problem. Snagtrail groups repeats by where they happen in your code, so your phone buzzes once and the count does the rest.
Start freeSource maps turn minified noise back into your files. Your own code comes first, library frames step back, and the commit that changed the line sits right beside it.
Read the docsEach one shows up as a single issue with a count, never as a flood.
Uncaught errors in the browser and on the server, with the user's last few clicks.
Requests that cross your limit get flagged before a customer has to tell you.
Background jobs that die quietly get retried, and reported if they fail again.
See the deploy that started the spike, and the deploy that ended it.
Add the package, set your key, and the next error you ship shows up with its line and commit.
$ npm install snagtrail import { trail } from "snagtrail"; trail.init({ key: process.env.SNAGTRAIL_KEY });
Add the whole team. Drag to your monthly error count and see what you'd pay. A quiet month costs nothing.
Over your limit? Snagtrail keeps the first event of every new issue, so nothing new slips past.
The SDK sends errors in the background, batched, after the response has gone out. Most apps see no change at all.
No, but you'll want them. Snagtrail uploads them on every deploy, so traces point at your real files instead of minified ones.
Snagtrail keeps the first event of every new issue, so nothing new slips past. Repeats are counted, not stored.
Yes. Pricing is by errors, not seats, so adding a teammate never costs anything.