Feature Request: Remote approval for op CLI / desktop prompts to support Agentic Workflows

Hey team,

Claude Code just release a new https://code.claude.com/docs/en/remote-control feature that lets you send prompts and commands to your desktop (or remote sandbox) directly from a mobile app. You might already be using it

I have a feeling a lot of devs are going to run into the same need for "remote 1Password auth" because of this.

I really love how smooth 1Password handles secret and env management. I keep all my dev keys in 1P, and a lot of my scripts are tightly coupled with the op CLI. For pure local development, this setup is flawless.

I tossed Claude Code into a local Docker sandbox for security reasons. Whenever it needed a 1Password key (like for git SSH), I just used SSH forwarding to my host machine for biometric auth. That worked fine since I was sitting right there.

But this new mobile remote control feature completely breaks that workflow. I'm not at my desk to scan my fingerprint. Sure, we could generate static tokens to let the agent grab what it needs, but that’s way too inflexible—agents need dynamic access to different resources.

If we could get a feature to approve desktop/CLI secret requests directly from the 1Password mobile app (maybe with some specific allow/deny rules and auth timeouts), it would be a massive help for agent-driven development. This applies to both local remote control and fully remote sandboxes.

Honestly, this would be a killer feature for me. Would love to know if this is something the team could consider.