Feature request for product/service
Chat
Describe the request
It would be convenient to have checkpoints before each Subagent invocation.
This way, if the Agent is performing a long task using multiple Subagents, and one of them breaks everything, you can simply roll back to the moment it was launched. You can also avoid forcing the Agent to commit after each Subagent invocation.
For a multi-Subagent call, I suggest creating a checkpoint at the moment the group is called, or tracking changes to each Subagent in the group.