I’m attempting to build a knowledge base for our Product Design team that can crawl our Figma files and getting authentication failed events with the recommendation of to ensure the XPath selector is valid and the element exists on the page. Any guidance would be much appreciated!
Hi @fedeeliz and welcome to the community!
I believe that error means your crawler reached the figma login page but couldn’t find the fields at the XPaths you provided, so it never completed the login. As per my understanding, figma organization accounts don’t use a conventional username and password form. They authenticate through SAML or SSO, so there are no stable fields for the crawler to target and the XPath step fails. Here’s the Web Crawler integration guide which also has troubleshooting steps in case they’re useful.
It would help to know how your team signs into Figma, email/password, Google, or SSO. That’ll tell us whether the XPath approach can work at all.
Hope this helps and feel free to reply with clarifications/questions about this topic!
Thanks for the quick response (and the warm welcome!). You’re absolutely correct - Figma uses SSO so this will all go through Midway for authentication. Does that mean web crawling isn’t possible for Figma?
Just wanted to follow up on this - thanks in advance!
Hi @fedeeliz,
Thanks for the follow-up! I did some more investigating into this matter, and unfortunately, yes, if your Figma authenticates through Midway, the web crawler won’t be able to handle that.
The crawler’s authentication options (form-based and SAML XPath) rely on finding username/password fields on a login page, and Midway doesn’t present those in a way the crawler can interact with. So web crawling authenticated Figma content isn’t going to work in this setup.
I wasn’t able to find a workaround at this moment, so I will go ahead and mark this post as a ‘feature-request’ and archive it to promote visibility to the product team. Thank you for bringing this up!