It is possible to create a new custom model node to use in an agent?

Questions
,
1

Describe the problem/error/question

Is it possible to create a custom model node for LangChain in n8n?
I’m trying to use Snowflake’s Cortex model in my agent. I created a custom LLM implementation and a node named LmSnowflakeCortex.node.ts under /packages/@n8n/nodes-langchain/. However, I’m not seeing the Snowflake Cortex node appear in the model list when running the editor locally.

What is the error message (if any)?

There’s no error message – the node simply doesn’t show up in the model selection list.

Please share your workflow

Not applicable – I’m currently working on node development.

Share the output returned by the last node

N/A – the issue is with the node not being recognized or listed.

Information on your n8n setup

  • n8n version: Latest (cloned from GitHub repo)
  • Database (default: SQLite): Default (SQLite)
  • n8n EXECUTIONS_PROCESS setting (default: own, main): own
  • Running n8n via (Docker, npm, n8n cloud, desktop app): pnpm dev (local development)
  • Operating system: macOS

Additional context

  • I also created a credentials file at /nodes-langchain/credentials/SnowflakeCortexApi.credentials.ts.
  • I added both the credentials and node paths to /nodes-langchain/package.json:
"n8n": {
  "n8nNodesApiVersion": 1,
  "credentials": [
    "dist/credentials/SnowflakeCortexApi.credentials.js"
  ],
  "nodes": [
    "dist/nodes/llms/LMSnowflakeCortex/LmSnowflakeCortex.node.js"
  ]
}

Thanks in advance! Any ideas why the node might not be showing up?

2

you might have to reference it somewhere, I would check another existing, and all files where its referenced, u can use grep or in vs maybe something like to package.json