Add some more context to the Touch ID modal

Question

The above is the modal I got when asking Claude CLI (and giv “him” permissions to do so) to invoke this command:

op read "op://rgbunhumnzulmueaiee2yifbde/vpooj5dlrzeexf2uf3potehwvm/token"

In the world of AI and multiple process doing parallel work, it would be great for security if you could provide more and better context as to who requested access to what.

In the above example:

Ghostty is the terminal software used. It would be great, if possible, to instead have the name of the program for the process that invoked the given comman.
It would also be great if you could print some info on what command triggered this, e.g. `op` or even `op read`, `gh` etc.

pelmered · Answer

Yes, I have thought about this a lot for the last months as well. This becoming an increasingly serious problem with both all the agents and the recent supply chain attacks. Each of those would look exactly the same, and I have no idea what I am giving access to with this prompt. We need to see exactly what password/item is being asked about and as much additional context as possible(maybe under an accordian?).

Useful things to should would be things like the full command, and if possible what process/agent inside the terminal asked for this (is it an for example an agent, npm run *, or something else?).

I hope you can prioritize this, because it feels like it is just a matter of time before I either get hacked, my keys leak or my agent does something really stupid with any of the credentials in 1Password.

Sign up

1Password employees: Sign in with SSO

Login to the community

1Password employees: Sign in with SSO