Memory node

I need to use a memory for whatsApp AI Agent. The simple memory is not good enough, could anyone tell me what is the best memory I can use or if there is a way to build long memory without using API?
I hope I find solution or video explaining that.

Describe the problem/error/question

What is the error message (if any)?

Please share your workflow

(Select the nodes on your canvas and use the keyboard shortcuts CMD+C/CTRL+C and CMD+V/CTRL+V to copy and paste the workflow.)

Share the output returned by the last node

Information on your n8n setup

  • n8n version:
  • Database (default: SQLite):
  • n8n EXECUTIONS_PROCESS setting (default: own, main):
  • Running n8n via (Docker, npm, n8n cloud, desktop app):
  • Operating system:

Hi @Abdullah.F.Al-Harazi

What do you mean by “not good enough”?!
What are your expectations that you couldn’t achieve with the memory in general?

Hi @mohamed3nan

I mean the simple memory most of the time forgets the user name and ask the same question again, and doesn’t remember the conversation. It is very weak and is not for serious workflow. I hope you can help please.

increase the context window length:

image

Thanks @mohamed3nan . I tried that and raise the context length to 100, but the agent assistant still ask for the same information already given. Take this scenario for example: (For the purpose of registration in training course, the assistant ask the user to provide name, email, company and other details. Then, the user provides all required details to complete the registration process. Then the user ask the assistant if he can contact with sales for contracting B2B. The assistant ask the user to tell his name and provide phone number and email …etc. in order the sales will contact the user for B2B).

See, this is one example of the problem I faced. The agent ask again and again for details he already got from the user.

I appreciate your help.