n8n Multi-Session OpenAI Chat Agent with Data Table Memory

Keep an OpenAI chat running across sessions with structured “memory”

This n8n workflow is an n8n Multi-Session OpenAI Chat Agent with Data Table Memory that uses a data table to persist information between conversations, so your agent can reliably recall prior context and act consistently over time.

What this workflow does

• Receives chat input and processes it through conditional logic using the if node.
• Stores and retrieves session context using a Data Table (via the Data Table and Data Table Tool nodes), enabling multi-session “memory.”
• Guides the agent with a visible working spec using a Sticky Note node for clearer workflow maintainability.
• Runs an OpenAI-capable agent through the n8nn8n-nodes-langchainagent node, leveraging the stored table data as part of the agent’s operation.

Use cases

• Customer support assistants in which each user/session retains prior conversation details.
• SaaS operations copilots that remember user preferences or prior actions stored in a Data Table.
• Automation engineers building reliable agent flows where stateless requests need persistent context.
• Internal knowledge Q&A where you maintain per-session or per-user memory in a structured format.

Technical details

• Nodes/logic: if, set, data table, sticky note, data table tool, n8nn8n-nodes-langchainagent
• Core capability: Multi-session memory implemented via Data Table read/write operations
• Agent framework: LangChain Agent provided by n8nn8n-nodes-langchainagent