What I'm proposing
Add support for specifying a given user as the owner/creator of a given action or otherwise support the ability to authenticate multiple user profiles.
So for instance, if I want to create a new message but post it as a specific user (such as a specific agent on the team) I could do something like:
basecamp todo "Fix bug" --in 12345 --user myagent@mydomain.com --to otheragent@mydomain.com
What problem this solves
If you're running an autonomous agent gateway setup (i.e. OpenClaw, Hermes, or similar) you may end up creating a team of agents with given specializations. If the agents want to use the cli (which has great ergonomics for agents) everything they create looks like it was made by the same user (the one who authenticates the CLI for them).
If the CLI allowed you to specify which user was taking the action in question, that would allow the agents to take actions in Basecamp and have those tied back to their specific user.
Thoughts/Notes
My thinking is that an autonomous agent team ultimately has similar needs/concerns to a fully human team where you need to know the status of work and get a sense of how things are coming along. The CLI itself and the agent skills are awesome (thanks for providing these) but it seems like it's currently tailored more for a single-agent setup. Basecamp itself is a great collaboration tool which feels like a natural fit for these types of agentic teams.
I'm happy to provide any other useful details or even contribute to this if I can in some way.
What I'm proposing
Add support for specifying a given user as the owner/creator of a given action or otherwise support the ability to authenticate multiple user profiles.
So for instance, if I want to create a new message but post it as a specific user (such as a specific agent on the team) I could do something like:
What problem this solves
If you're running an autonomous agent gateway setup (i.e. OpenClaw, Hermes, or similar) you may end up creating a team of agents with given specializations. If the agents want to use the cli (which has great ergonomics for agents) everything they create looks like it was made by the same user (the one who authenticates the CLI for them).
If the CLI allowed you to specify which user was taking the action in question, that would allow the agents to take actions in Basecamp and have those tied back to their specific user.
Thoughts/Notes
My thinking is that an autonomous agent team ultimately has similar needs/concerns to a fully human team where you need to know the status of work and get a sense of how things are coming along. The CLI itself and the agent skills are awesome (thanks for providing these) but it seems like it's currently tailored more for a single-agent setup. Basecamp itself is a great collaboration tool which feels like a natural fit for these types of agentic teams.
I'm happy to provide any other useful details or even contribute to this if I can in some way.