Repository navigation
[Docs]: How to structure Memory (chat session, across chat session) #644
Description
Activity
It's actually linked to #645
You can use the combination of org id and project id to separate the memory. If you need session id, you can put it in the metadata. We should be able to filter with the metadata.
@ChristianKniep did you resolve your issue? If so, let's close this issue. Thanks.
@ChristianKniep - There are no updates from you on this issue since December 1, 2025. I will close this issue on or after February 18th unless you provide a request to keep it open and an explanation.
This issue has been automatically marked as stale because it has not had recent activity. It will be closed in 14 days if no further activity occurs. If this issue is still relevant, please comment to keep it open, or add the
keep-openlabel if it should remain open indefinitely (e.g., a roadmap item awaiting a champion). Thank you for your contributions.
Documentation Page URL
https://docs.memmachine.ai/api-reference/search-memories
Describe the issue
I am working on the n8n community node and I am wondering how to structure the memory storage and retrieval.
In v1 I used a session ID of an ongoing session to retrieve the memories for the on-going conversation.
Within v2 the sessionID was removed. I wonder how I do that in v2?
Suggested change or fix
Session ID becomes ProjectID
I am using the sessionID from n8n (which changes for each chat session) as a projectID. In that way I am going to separate each session as a dedicated project.
Within the session I can keep the LTM and STM as context to the actual session.
Semantic Memory
I suppose I'll search for semantic memory for the group or even for the user. Those are shared across all sessions and become more or less the ground truth for the user...
Additional context
No response