Folder API legalty question

[luma98]

Can I legally, personal-use only, automate the folder browser API from an external program?

I need to upload my local library into Spotify. However, my files are structured on folders/playlists/songs, not just playlists/songs as the official WebAPI exposes.

By running a browser in debugging mode while attaching via Python's CDP playwright, I think I'm able to run javascript in the console and create folders in an automated way, thus being able to upload my local library into spotify and once and for all get rid of Winamp. For tracks and playlists, I use the official WebAPI. (I also use Gemini for track name and artist resolution/ best track pick for Spotify).

My question is, do you think this is legal?  I've read through Spotify's Developer Terms and the only thing I'm worried about is this one:
https://developer.spotify.com/terms#section-iv-restrictions
Section IV.2.a.ii "modifying, editing, altering, creating derivative works, disassembling, decompiling, reverse-engineering, or extracting source code from the Spotify Platform (including any client libraries), Spotify Service, or Spotify Content (except to the extent such restrictions are expressly prohibited by law). You may adjust the size of metadata or cover art as necessary;"

The thing is, I'm not reverse engineering anything, I'm running Js which is already in my browser, and I'm definitely not trying to reverse engineer their platform, I just want to listen to nice flacs on-demand instead of shitty 128kbps I downloaded in high-school 10 years ago.

[Cdrow]

I’m working on something similar and have also tested related Playwright automation across a few platforms without issues so far. If you’re just using standard browser automation and not doing anything high-frequency or large-scale (like bulk scraping), it generally behaves like normal user interaction from a technical standpoint.

That said, I don’t have legal expertise, so I can’t give a definitive interpretation of the Spotify terms. Even if it’s “just a browser,” services can still treat automated interactions differently depending on their policies.

As for that specific clause you mentioned, I don’t think your issue would likely come from there in this case. If anything, concerns would probably show up elsewhere in the ToS depending on how the automation is used, especially around automated access patterns or non-standard client behavior.

Practically speaking, if it stays low-volume, user-driven, and behaves like a normal session (just helping manage folders rather than mass extraction or manipulation), that’s usually where these kinds of tools tend to sit safely in practice.