What n8n already keeps

n8n’s chat memory sub-nodes keep a window of recent messages for a session key. They are a conversation buffer, not a store of facts with history.

Where Auld fits

Use two HTTP Request nodes — one before the AI Agent that recalls for the incoming message, and one after that ingests the turn — or give the agent an HTTP Request tool. Keep a small chat memory window for conversational flow.

Concernn8nAuld
Current conversationYesNot needed
Facts across sessions and channelsYour codeYes
Which fact is current after a changeYour codeYes, with history
What was true on a dateYour codeYes (asOf)
Who may see a memoryYour codeIdentities and audiences
Source for each statementYour codeReturned with every memory

Implementation sequence

  1. Store credentials

    Create a Header Auth credential with Authorization: Bearer ….

  2. Recall before the agent

    HTTP Request node posting to /v1/recall with the session’s user id.

  3. Pass memories in the prompt

    Reference the recall output in the agent’s system message.

  4. Ingest after

    Second HTTP Request node posting the message to /v1/records.

  5. Branch on results

    An IF node can route to a human when no memory exists for an account.

Example

HTTP Request node — recall
POST  {{ $env.AULD_API_BASE }}/v1/recall
Headers
  Authorization: Bearer {{ $env.AULD_API_KEY }}
  Content-Type:  application/json
Body (JSON)
  {
    "query":  "{{ $json.message }}",
    "viewer": "{{ $json.userId }}",
    "maxTokens": 2000
  }

Illustrative example · standard HTTP · check the framework’s current docs for exact signatures

Or connect the MCP server

If your n8n setup can use MCP tools, point it at the Auld MCP endpoint instead of writing wrappers. The model gets remember, recall, forget and memory_status. Use wrappers when you need to inject the user id yourself or post-process results; use MCP for speed of setup.

Connect an MCP client →

Custom integration recipe using standard HTTP/MCP tools

  • This recipe wires n8n’s own tool interface to the Auld REST API or MCP server; there is nothing Auld-specific to install.
  • Bind the viewer id from your authentication layer, not from model output.
  • Ingestion is asynchronous — poll the job before depending on a just-sent fact in tests.
  • Recall returns up to your plan’s token limit (8,000 on Pay as you go and Flex).

Frequently asked questions

Can webhooks trigger ingestion?

Yes. A Webhook trigger node can forward CRM or helpdesk events straight to /v1/records.

Self-hosted n8n?

Works the same; the instance needs outbound HTTPS to your Auld host.