TPM keys and user presence

[Martin Paljak]

> On 5. Jul 2025, at 17:42, James Bottomley via openssh-unix-dev <openssh-unix-dev at mindrot.org> wrote:
> 
> On Sat, 2025-07-05 at 10:55 +0200, Anton Khirnov wrote:
>> Quoting Brian Candler (2025-07-05 09:39:13)
>>> On 05/07/2025 06:52, Anton Khirnov wrote:
>>>> - I have to enter the PIN on every use, which is highly
>>>> inconvenient and
>>>>    increases the likelihood the entry will be observed (e.g. in a
>>>> public environment with cameras)
>>>> - the key is in the agent and PIN is not required, then any
>>>> program that can access the agent can silently SSH all it wants
>>> 
>>> If the private key is actually in the agent, then you can flag that
>>> key to require confirmation on each use. With a normal key, it
>>> would be "ssh-add -c". That normally just means clicking "OK"; you
>>> don't have to enter a PIN.
>> 
>> Thank you, that seems to be exactly what I was looking for.
> 
> The ssh community has rejected many approaches to TPM based keys, so
> the easiest way to use them is to use gpg-agent (for any 2.4 and up
> version of gpg) as the ssh agent backend and then simply use the gpg
> keytotpm command on keys you want to become only TPM accessible.

Respectfully disagree. You're free to run any ssh-agent, and it makes various hardware elements play along pretty decently (like secretive for mac).

For TPM2, see here:

https://github.com/Foxboron/ssh-tpm-agent

Also have a look here: https://fosdem.org/2025/schedule/event/fosdem-2025-5544-hardware-backed-ssh-keys-ssh-tpm-agent/


I'd only recommend going down the GPG path if you're already established and invested in the GPG infrastructure/setup, but never for beginners.


Cheers,
Martin