This question/request is related to DeepAgents, and I hope I’m posting in the correct category.
I really like that DeepAgents try to emulate the workflow of common agentic tools like Claude Code and Cursor. So far, my experience has been very positive. One feature that exists in these tools but seems to be missing here is the mechanism they use to ask clarifying questions or gather additional context from the client. Often, when a prompt is unclear, these tools return a set of multiple-choice questions for the user to answer in order to better understand the request.
I believe this is an essential part of a DeepAgent’s workflow and should be included as a built-in capability (tool).
What do you all think?
I would be happy to contribute towards implementing this feature.
Hi @MinuraPunchihewa
Yes — I completely agree
Clarification shouldn’t rely on the model “remembering to ask.” It should be a built-in workflow step in DeepAgents.
Tools like Claude Code and Cursor formalize ambiguity handling with structured, often multiple-choice questions. That reduces hallucinations and prevents wrong execution.
Making clarification a first-class tool (e.g., request_clarification() that pauses execution until resolved) would improve safety, determinism, and reliability.
Hi @MinuraPunchihewa if it’s solved, a huge favor - please mark this post as Solved for the others, so that they can get benefits from it 
Well, I am happy to mark it as resolved, but is there a plan to add this to the future roadmap?
I am also hoping for the future roadmap on this @MinuraPunchihewa