Atom feed encoding issue (UTF-8 declared but garbled content)

Hi,

I noticed a possible encoding issue in the Atom feed.

Example feed:

https://status.cafe/users/xxx.atom

The feed declares UTF-8:

<?xml version="1.0" encoding="UTF-8"?>

However, some entries contain garbled text (example):

馃槑 璁哄揩...

The same content renders correctly on the HTML page, so the issue seems to occur during feed generation rather than client-side parsing.

This appears to affect non-ASCII characters (e.g. Chinese and Japanese text), which are rendered as garbled text in the Atom feed.

I also verified this is not caused by client-side decoding.

Questions:

1. Is this a known issue with the Atom feed?

2. Is the feed internally generated with a different encoding?

3. Is there any plan to fix or improve this?

Thanks!