Hi!
I’m developing an app to help developer with worklog. they have a list of tasks planned for a day.
Agent need to read what was planned for the day and asked developer about each of them in order to get the status. And then another agent stored it back.
The problem is to how make agent ask questions one by one for each task separately? “human_input=True” only implies one reply
This is my understanding: If you make sure that a task/agent is in the input loop of your process then on each iteration the Agent should prompt for human input.
E.G.:
You have a Task which uses an external tool to get a list of jobs from a csv file.
Task: Get jobs from list and with each one do ‘some functional description’
The Agent that is assigned to the task will have & be recognized as the functionality as described in the ‘functional description’
If I’m wrong in my assumptions let me know.