Incomplete checkpoint design limits rollback and message editing #2348
Labels
Comments
|
200% agree. It's the only substantial problem I have with Roo Code at the moment. |
|
It’s been a while since I looked at the checkpoint code. Do you have a good sense for how it would need to change? |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
App Version
3.11.7
API Provider
OpenRouter
Model Used
N/A
Actual vs. Expected Behavior
The current checkpoint feature is not fully developed. At the moment, checkpoints are created only occasionally—typically when Roo sends a reply.
However, in real-world usage, users often expect to roll back a task to the state before they sent a message, allowing them to revise or rephrase their input. This is a common behavior pattern in many web-based AI chat interfaces.
Currently, users cannot directly edit previous messages, nor can they restore the state just before a specific message was sent. This makes it difficult to iteratively refine prompts or explore alternate directions without starting the task over from scratch.
Detailed Steps to Reproduce
Relevant API Request Output
Additional Context
No response
The text was updated successfully, but these errors were encountered: