[FEATURE] Feature request: allow claude to never stop working on a prompt. `/never-stop <prompt>`

Preflight Checklist

• [x] I have searched existing requests and this feature hasn't been requested yet
• [x] This is a single feature request (not multiple features)

Problem Statement

claude keeps giving control back to the user. there's a subset of tasks where I never want control back, just keep trying forever claude

Proposed Solution

I'd like a mode where Claude Code keeps working toward a goal indefinitely, only stopping when the user manually intervenes.

you can imagine a prompt like:

/never-stop Fine-tune this model until it beats gpt-5 on MATH benchmark, when that happens keep making model improvements forever

Alternative Solutions

_No response_

Priority

High - Significant impact on productivity

Feature Category

CLI commands and flags

Use Case Example

_No response_

Additional Context

_No response_