Description
First off, thanks for your work on this awesome project! It's been fun to work with.
I'd like to propose some changes to the documentation and naming of variables to make things a little more clear. (I'll save the variable names topic for a separate issue). These proposed changes come from my experience in exploring the docs and the code.
I'm happy to submit a pull request on this myself. The jist of it is the section on calling agents could use more details and some broader context so it makes sense. Specifically, I'd like to add more to https://github.com/evalstate/fast-agent/tree/main?tab=readme-ov-file#calling-agents
The critical details are in comments and without broader context it's unclear what they mean. Id like to add a section explaining how agents are registered, what happens during the setup in the fast.run() context manager, and make it more clear about what is bound to the agent variable.
I think doing this would make it more clear how the different methods of interaction with the agent object work. I wracked my brain for some time trying to understand what the correct approach was especially since my use case (using in a web server) was different from the one focused on in many of the examples (using in interactive mode on cli).
Again, thanks for all the work you've done. Like I said I'm happy to make a pull request to add details to the docs. But either way, I hope this feedback is useful for you!