Ask or Assume? Uncertainty-Aware Clarification-Seeking in Coding Agents

TL;DR

This paper introduces an uncertainty-aware multi-agent system for coding agents that effectively detects underspecification and asks clarifying questions, significantly improving task resolution rates in software engineering tasks.

Contribution

It presents a novel multi-agent scaffold that separates underspecification detection from code execution, enhancing clarification-seeking capabilities of LLM agents.

Findings

Multi-agent system achieves 69.40% task resolve rate, outperforming single-agent systems.

The system exhibits well-calibrated uncertainty, balancing queries based on task complexity.

Proactive clarification improves agent collaboration in underspecified tasks.

Abstract

As Large Language Model (LLM) agents are increasingly deployed in open-ended domains like software engineering, they frequently encounter underspecified instructions that lack crucial context. While human developers naturally resolve underspecification by asking clarifying questions, current agents are largely optimized for autonomous execution. In this work, we systematically evaluate the clarification-seeking abilities of LLM agents on an underspecified variant of SWE-bench Verified. We propose an uncertainty-aware multi-agent scaffold that explicitly decouples underspecification detection from code execution. Our results demonstrate that this multi-agent system using OpenHands + Claude Sonnet 4.5 achieves a 69.40% task resolve rate, significantly outperforming a standard single-agent setup (61.20%) and closing the performance gap with agents operating on fully specified instructions. Furthermore, we find that the multi-agent system exhibits well-calibrated uncertainty, conserving queries on simple tasks while proactively seeking information on more complex issues. These findings indicate that current models can be turned into proactive collaborators, where agents independently recognize when to ask questions to elicit missing information in real-world, underspecified tasks.