[Question] 是否支持终端任务的RL训练（类似Terminal Bench）

Your Question

Does this framework currently support (or have plans to support) Reinforcement Learning training for terminal-based interactive tasks?

Specifically, I am interested in environments similar to Terminal Bench, where the agent needs to:

Interact with a bash shell or CLI tools.

Handle multi-step decision-making based on terminal outputs.

Learn from rewards derived from task completion (e.g., successful file manipulation or system configuration).

Given the current search-r1 and slime-mmq examples, is there a recommended way to wrap a terminal environment for training within this repository?

What I've Tried

Only read the Quick Start Guide but not found

Environment (if relevant)

• slime version:
• Python version:
• PyTorch version:
• CUDA/ROCm version:
• GPU type and count:
• OS:

Additional Context

No response

Pre-submission Checklist

• I have read the CONTRIBUTING.md and understand the collaboration scope.
• I have read the documentation and FAQ and my question is not answered there.
• I have searched for existing issues and my question has not been asked before.

Start by reviewing the Quick Start Guide, the search-r1 and slime-mmq examples, and the linked documentation and FAQ. Determine whether they expose an existing way to wrap interactive terminal tasks; done would require a maintainer-confirmed support path or a documented plan for adding one.