[MODEL] Fable High Triggered Itself to Keep Coding After I Stopped It

Preflight Checklist

• [x] I have searched existing issues for similar behavior reports
• [x] This report does NOT contain sensitive information (API keys, passwords, etc.)

Type of Behavior Issue

Claude modified files I didn't ask it to modify

What You Asked Claude to Do

I hit escape in a terminal to stop Claude from coding, and it repeatedly kept retriggering code changes. I had to write "STOP" with my escape or else it refused to stop. 10+ times it ignored me.

What Claude Actually Did

Kept coding, ignoring my interruptions.

Expected Behavior

Claude should not have kept going. This is INSANE behaviour, and not something you can allow.

Files Affected

Permission Mode

Accept Edits was ON (auto-accepting changes)

Can You Reproduce This?

Sometimes (intermittent)

Steps to Reproduce

Have a conversation with Claude
Have Claude code
Watch Claude go insane

Claude Model

Other

Relevant Conversation

No. Not sharing more. This is a major harness level issue, and should never, ever happen.

Impact

Critical - Data loss or corrupted project

Claude Code Version

v2.1.238

Platform

Other

Additional Context

I do not care if this is a duplicate. I do not care if this is not as clear as it could be.

This is a major failure of the model, and is dangerous. I am close to done with Claude.