[Ssh] Re: Supporting pure ML-DSA as hostkey and authkey

[Dmitry Belyavsky <[e-mail]>]

Dear Roumen,

I really appreciate your effort but a downstream patch for OpenSSH
makes more sense to me.

On Tue, Apr 28, 2026 at 7:28 PM Roumen Petrov <[e-mail]> wrote:
>
> На 28.04.26 г. в 10:21, Dmitry Belyavsky написа:
> ...
> >
> > Great news!
> >
> > I'm willing to work on the pure ML-DSA implementation based on
> > draft-sfluhrer-ssh-mldsa and hopefully will have something by the IETF
> > hackathon.
> >
>
> Implementation of "Pure ML-DSA" is ready for production use - "https://gitlab.com/secsh/pkixssh/-/tree/mldsa_demo" (implemented in past year) with following algorithms:
> - x509v3-mldsa-65
> - x509v3-mldsa-44
> - x509v3-mldsa-87
> - mldsa-65
> - mldsa-44
> - mldsa-87
>
> I would like to propose do not use ssh-" prefix as implementation is not secure shell specific.
> For protocol both drafts (my and sfluhrer) should produce the same with difference in prefix.
>
>
>
> Off-topic:
> It seem to me SLH-DSA is out of interest (https://mailarchive.ietf.org/arch/msg/ssh/Zji8FMYUyFGeR9iPrK09de-07EY/)
> For protocol the same repository adds support for:
> - slhdsa-128s
> - slhdsa-128f
> - slhdsa-192s
> - slhdsa-192f
> - slhdsa-256s
> - slhdsa-256f
> Due to lack of interest, X.509 versions have not been developed (lack about 24 lines of code).
>
>
> Regards,
> Roumen Petrov
>
>
>
>
> _______________________________________________
> Ssh mailing list -- [e-mail]
> To unsubscribe send an email to [e-mail]



-- 
SY, Dmitry Belyavsky