Open truekonrads opened 1 year ago
Is it duplicate of #172?
Do you have any additional confirmation (besides scanner results) that it is not a false positive?
Note that I have attached additional file "websocat_rebuild.x86_64-pc-windows-gnu.exe" which shows less (though nonzero) detection entries on VirusTotal.
Hi - same issue. I did a basic check to see if if the triggers are mostly on "socat" and the answer to that is yes. If I replace in the binary "socat" with something else and "dest-unreach" wit hsomething else the results are only 13/72 vs original 38/72. I think if you want to provide windows builds you need to EV code sign the artefacts or rename your project away from the word "socat". Have a look here for some code signing stuff.
Some Windows builds of Websocat show no detections (example: https://www.virustotal.com/gui/file/a03ab5059a1250f88825315f9ee9297419d0104297f49a37bf3f2a3a6fc032f2).
rename your project away from the word "socat"
Socat is also a legitimate networking tool (and is a direct inspiration of Websocat). It should not be unconditionally detected as a malware, though malware, like other software, can use all those networking tools as components.
I don't think it's a good idea to condone to security scanners' false positives in such a drastic way as to rename the project.
EV code sign the artefacts
How do I do it (not being a Windows or Microsoft user)? Do you have a guide of Linux to Windows crosscompilation with code signing? Is EV code signing friendly towards Open Source? Should "Cosign" tool from the linked website provide protection from the false positives?
https://www.virustotal.com/gui/file/00db6782e44cdfe2351196911f038fc0e94e0eb48dcbdb3db88d1bbc71d89344