I appreciate how the models break large tasks into smaller, manageable steps. After each step, it prompts with “Do you want me to continue to…?”, and the confirmation sounds help signal progress. However, I find it repetitive to constantly monitor the process and manually type “continue” or “go ahead” each time.
Is there a way to automate this step or instruct the model to proceed through all steps without repeatedly asking for permission?
Note: I don’t mean permission to run tools. I mean, stuff like this:
You can now see a Divider frame in your app.
Ready to proceed with the next component (Dropdown)?
2 Likes
It can be automated, you need to set it up yourself, but this is not very convenient, it is almost impossible to fully automate complex tasks, and AI repeated modifications often can’t be fixed, it is best to atomize processing, verification at each stage, is really efficient, although it seems complicated.
In my scenario: Each task is simple and requires the same steps to complete. Success rate is very high. So I really don’t want to manually tell the agent to proceed.
Create a.tsx
Create b.tsx
Create c.tsx
Imagine the agent completing one task and asking for permission to go on. It’s repetitive.
Examples for illustration purposes only.