Feature request: ability to delete/erase domain data added to RAG workspace

[saragerion]

Summary

It could be useful to be able to selectively delete specific domain data added in the RAG workspace.
Domain data I am referring to includes:

• Files uploaded
• Text
• Q/A
• Websites

Motivation

There are some use cases in which a company may want or need to delete specific domain data after it has been added to the RAG storage. This can happen for example when companies need to adhere to compliance requirements and therefore should be able to erase private data from their workloads when requested.
Example: European Union’s General Data Protection Regulation (GDPR) “right to erasure” or “right to be forgotten” requirements, when applicable.

[ajaylamba-provar]

Or the more common use case would be - the content has been updated.

[azaylamba]

@flamingquaks @massi-ang I'm trying to make changes for this, could you guys give some pointers to start with please?

[azaylamba]

I am working on this one. I have successfully implemented this for OpenSearch but my fork is not in sync with main after migration to AppSync.
I will sync with main and will modify the code accordingly.

[azaylamba]

Update: Changes are done, testing in progress.

[github-actions]

This issue is stale because it has been open for 60 days with no activity.

[azaylamba]

Finally, raised PR for this.

[azaylamba]

@saragerion Changes for this have been merged to main, this issue can be closed as done.