Feel free to close this. It's a nice-to-have for me. The little dots besides both the agents and the projects currently have two colors. One indicates non-active and one indicates active. I wish it was richer. The distinctions that I would like to cover are:
- inactive/ready agent
- working agent
- completed agent (it was working, now it's complete, but you haven't looked at the results yet - once you look at the agent results it goes to the inactive color)
- agent error (something went wrong ... this might be challenging)
- dead agent (Maybe this is not necessary since it's so easy to restart an agent.)
I implemented this for Rook and it was kind of useful for getting a quick overview of who's ready for me to give more input. The completed agent is the killer thing I need. Agent error was also useful, but it was easier for me to build w/ Rook because I assumed that every agent spoke Agent Client Protocol and so every error status was easy to observe.
Feel free to close this. It's a nice-to-have for me. The little dots besides both the agents and the projects currently have two colors. One indicates non-active and one indicates active. I wish it was richer. The distinctions that I would like to cover are:
I implemented this for Rook and it was kind of useful for getting a quick overview of who's ready for me to give more input. The completed agent is the killer thing I need. Agent error was also useful, but it was easier for me to build w/ Rook because I assumed that every agent spoke Agent Client Protocol and so every error status was easy to observe.