You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
possible to expand this to Kensington Validity 06cb:0088 ? #34
But encountered a slightly different folder structure and that's about where my skills end.
The item is a Kensington Verimark USB fingerprint reader.
If you could determine if it is suitable for this project and guide me how to proceed, I'd be happy to help.
Folder structure of the driver is:
KensingtonFingerprintApplication.exe
KensingtonFingerprintApplication.lnk
synaBscAdapter66.dll
SynaCP66.dll
SynaFPCoInstaller.dll
SynaFPService.exe
synahidumdf.sys
SynaU2F.exe
synaumdf.cat
synaWudfBioUsb66.dll
synaWudfBioUsbKensProd.inf
synaWudfU2fhid.dll
synaWudfU2fhid.ini
WudfUpdate_01011.dll
Looking at the .inf file, it appears that the driver GUID is the same as for 0097 and 009a (96710705-B080-4B29-A3EC-B16935AE663A), which means that synaWudfBioUsb-sandbox approach might just work. Have you tried renaming synaWudfBioUsb66.dll to synaWudfBioUsb.dll and following the rest of synaWudfBioUsb-sandbox instructions? I know, it's not an easy task. I've tried to make synaWudfBioUsb-sandbox and wine setup to be a bit more user friendly but got bogged down by the Wine packaging and gave up.
Do you know if your sensor is Match-on-Sensor or Match-on-Host?
@uunicorn I'm about done writing up some instructions for building the patched Wine in a Debian container, happy to open a PR for it if you're interested.
I can absolutely try to rename the dll.
What about: Copy *.xpfwext file from the extracted windows driver to c:\windows\system32 inside youre new Wine prefix
There is no such file.
Well, fwext is just an overlay on top of firmware in ROM. Perhaps it is not needed for this device.
The .inf file has an empty [FWextCopy] section, which implies that it is expected. synaWudfBioUsb-sandbox may still work.
The sensor is claimed to be match-on-host.
This is unfortunate. python-validity lacks support for MoH devices at the moment.
I've provided a bit more details in this comment.
I apologise, I think I've gone and made a new thread, when I should have read this one.
Am I correct in interpreting this discussion as the :0088 device isn't going to work / may not ever work?
I apologise, I think I've gone and made a new thread, when I should have read this one.
Am I correct in interpreting this discussion as the :0088 device isn't going to work / may not ever work?
You should look for Validity90, that one is generally for MoH.
It is not that it may never work, it is just (unlike Validity90) there is no existing framework to do the fingerprint matching inside the python-validity code at the moment. You are more than welcome to contribute it though.
I apologise, I think I've gone and made a new thread, when I should have read this one.
Am I correct in interpreting this discussion as the :0088 device isn't going to work / may not ever work?
You should look for Validity90, that one is generally for MoH.
I thought it was all but abandoned several years ago?
About Validity90, I just want to add that I wouldn't call it abandoned. I think the project was making progress and had a lot of traction among users. Then the owner disappeared. No idea if something happened to him, or if he got tired of the project and made a new github identity. I believe something happened to him, or he would have transferred ownership of the project. 3v1n0 picked it up, but it doesn't seem to keep the same pace.
It's a clean-room implementation that covers enroll, verify, and PAM/login integration end-to-end on Linux. The driver auto-pairs with the device on first plug-in (no Windows extraction needed) and persists pairing material to the standard libfprint state directory.
06cb:0088 turned out to be a pre-Prometheus chip with a different protocol from the post-Prometheus 06cb:009a that python-validity targets (no fwext partition, different pre-TLS pairing). So it's effectively a separate device family that didn't fit cleanly under python-validity's code paths.
A PR adding a clear redirect to that driver in python-validity's usb.py + README is at #258.
Activity
uunicorn commentedon Aug 30, 2020
Looking at the .inf file, it appears that the driver GUID is the same as for
0097and009a(96710705-B080-4B29-A3EC-B16935AE663A), which means that synaWudfBioUsb-sandbox approach might just work. Have you tried renamingsynaWudfBioUsb66.dlltosynaWudfBioUsb.dlland following the rest ofsynaWudfBioUsb-sandboxinstructions? I know, it's not an easy task. I've tried to makesynaWudfBioUsb-sandboxandwinesetup to be a bit more user friendly but got bogged down by the Wine packaging and gave up.Do you know if your sensor is Match-on-Sensor or Match-on-Host?
oofnikj commentedon Sep 3, 2020
@uunicorn I'm about done writing up some instructions for building the patched Wine in a Debian container, happy to open a PR for it if you're interested.
jontis commentedon Sep 4, 2020
I can absolutely try to rename the dll.
What about:
Copy *.xpfwext file from the extracted windows driver to c:\windows\system32 inside youre new Wine prefix
There is no such file.
The sensor is claimed to be match-on-host.
uunicorn commentedon Sep 5, 2020
Well, fwext is just an overlay on top of firmware in ROM. Perhaps it is not needed for this device.
The .inf file has an empty
[FWextCopy]section, which implies that it is expected.synaWudfBioUsb-sandboxmay still work.This is unfortunate.
python-validitylacks support for MoH devices at the moment.I've provided a bit more details in this comment.
jontis commentedon Sep 5, 2020
Interesting. Thank you for explaining.
In that case, I might get it's younger sibling to experiment with adding support.
It announced to be match-in-sensor.
Kensington VeriMark IT Fingerprint Key
What do you think of this?
uunicorn commentedon Sep 5, 2020
I think we can definitely try. Probably worth raising another issue with a correct usb id.
jaxjexjox commentedon Jan 12, 2021
I apologise, I think I've gone and made a new thread, when I should have read this one.
Am I correct in interpreting this discussion as the :0088 device isn't going to work / may not ever work?
himekifee commentedon Jan 12, 2021
You should look for Validity90, that one is generally for MoH.
uunicorn commentedon Jan 12, 2021
@jaxjexjox
It is not that it may never work, it is just (unlike
Validity90) there is no existing framework to do the fingerprint matching inside thepython-validitycode at the moment. You are more than welcome to contribute it though.jaxjexjox commentedon Jan 13, 2021
I thought it was all but abandoned several years ago?
himekifee commentedon Jan 13, 2021
Well, yes, but that is the most relative project.
uunicorn commentedon Jan 13, 2021
I think @3v1n0 's branch of
fprintdis still alive: https://github.com/3v1n0/libfprintjaxjexjox commentedon Jan 13, 2021
Thanks for your hard work and replies, people like you make using free software awesome, appreciate it.
jontis commentedon Jan 13, 2021
About Validity90, I just want to add that I wouldn't call it abandoned. I think the project was making progress and had a lot of traction among users. Then the owner disappeared. No idea if something happened to him, or if he got tired of the project and made a new github identity. I believe something happened to him, or he would have transferred ownership of the project. 3v1n0 picked it up, but it doesn't seem to keep the same pace.
parkerlreed commentedon Apr 26, 2021
Found another one in the wild
ID 06cb:0089 Synaptics, Inc.
https://www.amazon.com/gp/product/B07D3N3QHJ/
Works with Windows Hello after it grabs a Synaptics driver.
visorcraft commentedon May 26, 2026
For anyone still watching: a working FLOSS
libfprintdriver for the Kensington VeriMark (06cb:0088) now exists at https://github.com/visorcraft/Kensington_VeriMark_06cb-0088.It's a clean-room implementation that covers enroll, verify, and PAM/login integration end-to-end on Linux. The driver auto-pairs with the device on first plug-in (no Windows extraction needed) and persists pairing material to the standard libfprint state directory.
06cb:0088turned out to be a pre-Prometheus chip with a different protocol from the post-Prometheus06cb:009athat python-validity targets (no fwext partition, different pre-TLS pairing). So it's effectively a separate device family that didn't fit cleanly under python-validity's code paths.A PR adding a clear redirect to that driver in python-validity's
usb.py+ README is at #258.3v1n0 commentedon Jun 5, 2026
I guess it should be proposed against https://gitlab.freedesktop.org/libfprint/libfprint/ instead