We understand that the new feature of 1.3.3 may be causing false positives in some antiviruses. But two days ago, we asking Microsoft to inspect the file of the LovelyComposer 1.3.3 and the false positives in Windows Defender have been fixed.  ( It looks a little suspicious to automatic machine judgment, but it actually means that there is no problem when examined accurately by an expert. I'm thinking about how to deal with the false positive problem. )

If you're using Windows defender, I don't know why you're still having issues, but try updating your definition files to the latest version. You can simply update from the security settings or have them try the steps in the image.

If that doesn't work, or if you are using another antivirus, please let us know its name and, if possible, the version of the definition file.

Thank you for your report.

I didn't know about antivirus software called MalwareBytes and Rogue Killer. It may be the result of those software reacting instead of Windows Defender.

Lovely Composer is developed in Python and includes a Python script as an official add-on sample. The add-on's source code is completely public and harmless, and Lovely Composer does not run external scripts without explicit user interaction. However, such aspects cannot be detected by the mechanical processing of anti-virus software, and will only warn that there is a possibility of suspicious behavior.

It seems that in modern software development, the more convenient and powerful functions are added for users, the more antiviruses are suspected of being abused. This tendency seems to be particularly strong in tools rather than simple games.

Basically, I'm a game developer, so I'm confused due to my lack of experience, but I want to prevent false positives as much as possible. Some of the options are expensive and may not be available with cheaper software. It may be necessary to remove the function once implemented depending on the case. It looks like it will take some trial and error for a while.