Robots.txt file for SvelteKit projects

Here’s a real quick one on making sure you don’t blow your crawl budget on your SvelteKit project JavaScript files.

I noticed this almost 3 years ago and didn’t write about it. Tangentially related, I’ve recently been having a bit of a mare with Google indexing things I don’t want it to. So now I’m sharing it with you to help if you’re having similar issues.

So, I first noticed this when I was working with the absolute SEO legend that is Dino Kukic! He clued me into finding where your crawl budget is going in the Google Search Console.

To check your Google crawl stats, from the Google search Console on the main page go to the ‘Settings’ section for the site you want to check.

Then click on ‘OPEN REPORT’ for the ‘Crawl Stats’ section and you’ll see the crawl requests breakdown.

In my case, the ‘By file type’ section was showing that I was using up a lot of my crawl budget on JavaScript files, like 30% of my crawl budget was being used on JavaScript files.

From what I can glean, JavaScript content for the Google bot isn’t primary content and HTML should be prioritized.

So, in my robots.txt I’ll add in a Disallow to stop the JavaScript files from being crawled.

# https://www.robotstxt.org/robotstxt.html
User-agent: *

Disallow: /*.js$
Disallow: /*.json
Disallow: /analytics.json
Disallow: /current-visitors.json

The asterisks allows any file name, the dollar sign ensures it only matches the end of an URL and not a oddly formatted url (e.g. /locations.json.html)

I’ve also added in a few other paths that I don’t want to be crawled like the analytics.json and current-visitors.json API endpoints.

If you go to the GitHub for this site and check out the robots.txt file you’ll also see a ton of additional Disallow rules that I’ve added in to stop Google from crawling things I don’t want it to.