[BUG] Claude code stops working on pre-agreed plans. It should go all the way and execute the full plan.

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?

Claude code stops working in pre-agreed plans. It doesn't display a prompt to me. There are no external or internal obstacles preventing ongoing work. It just stops working. Sometimes with a question like: "Do you want me to continue with the next step?" (this is not displayed as a user prompt, but just a question in the progress text that Claude code is sharing).

Then just saying it should continue - it does continue.

But nothing prevented it from continuing in the first place. In fact, we agreed on the plan and we agreed that it will execute on it all the way.

This way I cannot trust Claude code to execute on a plan but have to babysit and check in on a regular basis. It's just wasting a lot of time not doing anything, when there is nothing preveting it from completing the agreed work.

What Should Happen?

It should complete the work we agreed it will perform, all the way.

Error Messages/Logs

Steps to Reproduce

1. Agree on a plan, which has multiple steps
2. Tell Claude code to execute on the plan
3. See it stop at some step and not continue
4. Tell it to continue - and it will continue (but then it shouldn't have stopped)

Claude Model

Opus

Is this a regression?

No, this never worked

Last Working Version

_No response_

Claude Code Version

2.1.263 (Claude Code)

Platform

Anthropic API

Operating System

Ubuntu/Debian Linux

Terminal/Shell

Other

Additional Information

_No response_