Seems strange that the dev seems to be keeping quiet on this, no? I’m not telling you to read every comment of every post, you can just skim the post titles. Then you’ll see multiple open issues and a few closed issues too going back 5 days to the latest BtS update.
Though I haven’t followed this project long enough to tell if this is just the way they normally behave.
Edit:
I’m back at my computer, so it’s easier to edit and add info now.
Some key points that have stuck out to me:
-
Previous version released in July only triggers 2 detections on Windows defender versus 29 for the most recent version: https://i.imgur.com/GIoH7eG.png
-
Users getting constantly pestered to update to the latest version: https://i.imgur.com/Oege3kU.png
-
Yeah, naturally, the dev is going to say it’s a false positive. Obviously. I’ve only mentioned that the dev has previously responded because some people barely skimmed through the issues and thought the dev simply hadn’t seen the latest open issue from only a few hours ago, when that is not the case.
So, the “[edit:
last] previous update” was built fromac41318, since then there were exactly 2 commits:Both do not immediately look malicious. So, either the release is poisoned (in which case you can build it from source and see if still detected), or the repo was poisoned before, and the payload didn’t activate until those changes, or AVs decided to crackdown on random shit running their code in other law-abiding processes’ address space 🤣
It’s a dll injection. Of course it gets flagged as a virus, because technically it is. That doesn’t mean that it is malicious.
Here is an example… On paper, reshade is a horrifically dangerous piece of software. It doesn’t get flagged only because it is well known and virus scanners have an exception for it.
Any of these geniuses stopped to think that Spotify changing its code and altering the way that it interacts with the dll could result in more “detections”?
did you want to link #573 ? you only linked the issues list
Nah I linked the issues page on purpose since there are multiple posts talking about it
Looks like the previous version only had two positive hits on VirusTotal, according to comments, whereas this newest version has 29.
Some said the previous version is still available. I don’t really have skin in the game, so nobody should take my advice without doing your due diligence.
I wouldn’t download / update until this gets resolved. Or maybe look for alternatives, or compile/build it yourself. It doesn’t necessarily mean it’s the developers fault. Could be something else. But maybe don’t download something that might contain a Trojan.
Seems strange that the dev seems to be keeping quiet on this, no?
the issue was just posted 7 hours ago. maybe they just haven’t seen it yet.
someone in issue #573 asked if the dpapi file is really needed, and by looking at the manual installation instructions, yes, because that contains all the code.
the developer loads custom code into the spotify process by using such an “override” dll file. it works because spotify is voluntarily loading a dll with this name, and if there’s such a file in the directory besides the .exe file, it’ll take precedence over the original file installed in the system.
the trojan warning is probably triggered because this technique is often used by malware to change the behaviour of your programs, but as with most technologies, it has good uses toothe issue was just posted 7 hours ago. maybe they just haven’t seen it yet.
There are multiple posts going back 5 days of people asking about it. Check closed issues too, the dev even responded to some of them by saying it’s only a false positive.
the dev even responded to some of them by saying it’s only a false positive.
What else are you expecting them to do then if they already answered? Write an essay on DLL injection and walk everyone through the code line by line to convince them it’s not malicious?
In the end you either have to verify the code yourself or you have to trust them when they say it’s a false positive.
Not the whole code but only the part that triggers those flags. Not everyone is versed in C to “verify the code” himself… That’s a stupid take, It’s like saying to a toddler to change his diapers on his own when it’s dirty.
Strangely enough It went from 1 trigger to 29 triggers after 1 update? Seems rather sketchy :/ In the past (pirated games/software) I would have ignored those warnings and add an exception into my firewall… But nowadays with all the crypto schemes and obfuscated code, I won’t go near anything like that.
Not the whole code but only the part that triggers those flags. Not everyone is versed in C to “verify the code” himself…
You don’t say. And the developer they don’t trust pointing to some piece of code and telling those people who cannot understand it themselves that it’s not malicious achieves anything?
If it’s a false positive there isn’t even anything to show in the first place. Nobody but the antivirus vendors know for sure why something triggers a false positive.
That’s a stupid take, It’s like saying to a toddler to change his diapers on his own when it’s dirty.
It’s like a toddler telling you you’re changing their diapers wrong and expecting you to explain to them what you did wrong even though you did everything correctly and the toddler doesn’t know anything about changing diapers in the first place.
I guess it’s all a question of point of view and reference point. 💁 I can’t argue against your opinion on the other side.
I do agree though that from this point of view it also make sense.
What else are you expecting them to do then if they already answered? Write an essay on DLL injection and walk everyone through the code line by line to convince them it’s not malicious?
I said that to indicate that the dev had already responded to the posts, and they were not in a different time zone or on vacation, as you suggested in another comment.
Seems strange that the dev seems to be keeping quiet on this, no?
It’s only been a few hours since the issue (I assume you’re referring to) was opened. The developer could be in a different time zone or on a vacation and not respond for a few weeks. People are not entitled to a (quick) response.
Though I haven’t followed this project long enough to tell if this is just the way they normally behave.
Looks completely normal to me. The tool works by pretending to be some DLL loaded by Spotify, providing the same functionality as the original library, but also modifying Spotify’s behavior to block ads. It’s easy to see why anti virus software would flag a modified DLL, injecting unknown code as suspicious, especially if the same DLL might have previously been used by some malware to inject malicious code.
