Scrapy provides 403 error, nonetheless works on native

Hello,
I even possess wrote a spider and it's engaged on the entire.
I even possess do up USER_AGENT in settings.

Nevertheless after I deployed on PythonAnywhere now it provides me 403 error:

DEBUG: Crawled (403)  (referer: None)
DEBUG: Rule at line 12 without any user agent to position into stamp it on.

What might perchance or no longer it’s? What is this line 12?

UPDATE: tested on Heroku and it provides me 200. So whats up with PythonAnywhere?

UPDATE2: they most attention-grabbing enable walk domains: https://www.pythonanywhere.com/boards/matter/27574/