[BUG] In auto mode, AskUserQuestion auto-selects the recommended option after ~60s of no user input (started in v2.1.198)
Preflight Checklist
- [x] I have searched existing issues and this hasn't been reported yet
- [x] This is a single bug report (please file separate reports for different bugs)
- [x] I am using the latest version of Claude Code
What's Wrong?
On v2.1.198 with auto mode active, when Claude presents a multiple-choice question (AskUserQuestion) in plain conversation, not responding within ~60 seconds causes the agent to proceed with its own "(Recommended)" option instead of waiting.
This began today (same day as 2.1.198); I did not experience this on prior versions.
Actual Behavior
After ~60s the dialog resolves on its own and the agent continues with
its recommendation, taking actions without consent.
What Should Happen?
The question waits for explicit user input. If auto mode intentionally times out questions, it should say so, be configurable, and never treat silence as approval of an option.
Error Messages/Logs
Steps to Reproduce
- Start an interactive terminal session on v2.1.198, permission mode: Auto
- Prompt something that makes Claude ask a multiple-choice question
(plain conversation, no custom command or skill involved)
- Don't touch the keyboard for ~60 seconds
Claude Model
Opus
Is this a regression?
I don't know
Last Working Version
_No response_
Claude Code Version
2.1.198
Platform
Anthropic API
Operating System
macOS
Terminal/Shell
Terminal.app (macOS)
Additional Information
- #30740 requested a configurable/disable-able AskUserQuestion timeout —
closed as not planned. This report is different: the timeout now
selects an option and acts on it, and only in auto mode. In default
mode the same prompt waits indefinitely (cf. #70294), so silence being
treated as consent is an auto-mode-specific regression.
This issue has 2 comments on GitHub. Read the full discussion on GitHub ↗