Skip to main content
n8n has an MCP Client Tool node, so an n8n AI Agent reaches Renor the way Claude Code does: one MCP address and one key.

​
Set up

  1. In the dashboard, open API keys and choose Create key. It is shown once.
  2. In n8n, create a Header Auth credential named for example Renor: header name Authorization, value Bearer <your key>.
  3. Import integrations/n8n/renor-memory.workflow.json from the Renor repository (Workflows → Import from file, or n8n import:workflow --input=integrations/n8n/renor-memory.workflow.json).
  4. On the Renor memory node select the Renor credential; on Chat model select your model credential.
The workflow is a chat trigger, an AI Agent, a chat model and the MCP Client Tool node set to Streamable HTTP. The node offers the agent two of Renor’s 24 tools (Tools):
  • get_context: the agent calls it before answering, with the current request.
  • remember: the agent calls it when you ask it to keep something.
Offering only these two keeps each request small; remove the tool filter on the node to give the agent all 24.

​
Try it

In the chat, say “Remember that I deploy with Kamal to the server orca-7.” Start a new chat and ask “How do I deploy?” The second chat has no history, so the answer can only come from Renor. The key belongs to your workspace: whatever this agent saves, your other connected agents can recall (see Several agents, one memory).