Hi Filip, appreciate the response! > On Feb 18, 2026, at 8:27 AM, Filip Skokan <[email protected]> wrote: > > > The 64 byte recommendation > > The revised HPKE spec does not have such limits, it in fact says that the > usable space is effectively limitless and instead says that it may be useful > for some implementations to impose a lower limit on the values they will > accept and if so that such limit be no less than the maximum Nsk of an > implementation-supported KEM but does not make that a recommendation. > https://www.ietf.org/archive/id/draft-ietf-hpke-hpke-02.html#section-7.2.1-5
By “lower limit” I think it means an upper limit smaller than the effectively limitless size. (“lower limit” can all be interpreted as the minimum size, but I don’t think what is meant here). That text goes on to say 66 bytes is a good number. It’s not a should or recommendation, but it is there. Seems kind of unclear and messy to me, particularly because 66 bytes is really small. LL
_______________________________________________ COSE mailing list -- [email protected] To unsubscribe send an email to [email protected]
