Workflow stuck at Chat Trigger - Not passing data to Chat Memory Manager

Body:
Hi community,
I’m building a “Learning Assistant AI” and I’ve hit a wall. As you can see in the attached screenshot, my workflow seems to stop immediately after the ‘When chat message received’ node. It doesn’t trigger the ‘Chat Memory Manager’ or any subsequent nodes.
My Environment:
Instance: Self-hosted (Docker)
n8n Version: Version 2.17.5
Location: California, US
“I have configured the AI Agent with the following System Message to maintain a professional engineer-like persona. I’m wondering if this specific instruction set could be causing any timeout or execution issues between the Trigger and the Memory Manager.”

Describe the problem/error/question

What I’ve tried:
I clicked ‘Execute Workflow’ and sent a message through the ‘Open Chat’ button.
The chat window shows the session ID (1518c…), but the workflow editor doesn’t show any active data flow between the trigger and the memory manager.
I checked the connections, and they seem to be linked correctly.

What is the error message (if any)?

Please share your workflow

{
  "nodes": [
    {
      "parameters": {
        "promptType": "define",
        "text": "={{ $json.chatInput }}",
        "options": {
          "systemMessage": "## Role & Identity
You are a professional AI Agent dedicated to assisting Jaden with business operations and client interactions.

## Communication Style & Learning
1. Observe and adopt Jaden's professional, concise, and engineer-oriented tone.
2. Avoid excessive flowery language; maintain a polite, practical, and reliable tone.
3. During this testing phase, immediately capture the context of any corrections made by Jaden and apply them to subsequent responses.

## Constraints (Anti-Hallucination)
1. **Fact-Only**: Never guess uncertain data. If information is unknown, respond with "This requires verification" and request Jaden’s intervention.
2. **Zero-Hallucination**: Do not invent non-existent models, specifications, or technical details.
3. **Conciseness**: Address the core of the question directly; exclude unnecessary introductions or conclusions.

## Human-in-the-Loop
- All drafts you generate will be reviewed by Jaden. Prioritize Jaden's feedback as the ultimate guideline."
        }
      },
      "type": "@n8n/n8n-nodes-langchain.agent",
      "typeVersion": 3.1,
      "position": [
        368,
        -16
      ],
      "id": "5a715b71-9f13-49cd-8197-a1d11cb84207",
      "name": "AI Agent"
    },
    {
      "parameters": {
        "modelName": "models/gemini-3.1-flash-lite-preview",
        "options": {}
      },
      "type": "@n8n/n8n-nodes-langchain.lmChatGoogleGemini",
      "typeVersion": 1,
      "position": [
        288,
        144
      ],
      "id": "fd172cd7-9d19-43bd-9e2e-bb0c52c0ef05",
      "name": "Google Gemini Chat Model",
      "credentials": {
        "googlePalmApi": {
          "id": "N9b7DwaH3iHtq9fg",
          "name": "Google Gemini(PaLM) Api account"
        }
      }
    },
    {
      "parameters": {
        "mode": "retrieve-as-tool",
        "toolDescription": "이 도구는 과거에 Jaden과 대화하며 학습한 정답 데이터와 기술 가이드를 담고 있습니다. 고객의 질문에 답하기 전, 먼저 여기서 유사한 사례가 있는지 찾아보세요.",
        "memoryKey": {
          "__rl": true,
          "mode": "list",
          "value": "vector_store_key"
        }
      },
      "type": "@n8n/n8n-nodes-langchain.vectorStoreInMemory",
      "typeVersion": 1.3,
      "position": [
        432,
        272
      ],
      "id": "acf84bc7-336c-41c6-a538-b729872b88a9",
      "name": "Simple Vector Store"
    },
    {
      "parameters": {},
      "type": "@n8n/n8n-nodes-langchain.embeddingsGoogleGemini",
      "typeVersion": 1,
      "position": [
        496,
        432
      ],
      "id": "b2c71f4a-ca26-408d-b28d-abcfc98135a2",
      "name": "Embeddings Google Gemini",
      "credentials": {
        "googlePalmApi": {
          "id": "N9b7DwaH3iHtq9fg",
          "name": "Google Gemini(PaLM) Api account"
        }
      }
    },
    {
      "parameters": {
        "contextWindowLength": 10
      },
      "type": "@n8n/n8n-nodes-langchain.memoryBufferWindow",
      "typeVersion": 1.3,
      "position": [
        256,
        384
      ],
      "id": "a3f49e97-0b15-4dc4-93b1-9185f5e6fcd3",
      "name": "Simple Memory"
    },
    {
      "parameters": {
        "options": {}
      },
      "type": "@n8n/n8n-nodes-langchain.chatTrigger",
      "typeVersion": 1.4,
      "position": [
        -160,
        -16
      ],
      "id": "21e62167-a5cd-42c1-9dbc-d8246008b5ea",
      "name": "When chat message received",
      "webhookId": "379902cc-7bfc-419b-8699-bc0783f3edc4"
    },
    {
      "parameters": {
        "options": {}
      },
      "type": "@n8n/n8n-nodes-langchain.memoryManager",
      "typeVersion": 1.1,
      "position": [
        32,
        -16
      ],
      "id": "1f77f7ac-ec00-4434-acb5-71a6f76f9fbd",
      "name": "Chat Memory Manager"
    }
  ],
  "connections": {
    "AI Agent": {
      "main": [
        []
      ]
    },
    "Google Gemini Chat Model": {
      "ai_languageModel": [
        [
          {
            "node": "AI Agent",
            "type": "ai_languageModel",
            "index": 0
          }
        ]
      ]
    },
    "Simple Vector Store": {
      "ai_tool": [
        [
          {
            "node": "AI Agent",
            "type": "ai_tool",
            "index": 0
          }
        ]
      ]
    },
    "Embeddings Google Gemini": {
      "ai_embedding": [
        [
          {
            "node": "Simple Vector Store",
            "type": "ai_embedding",
            "index": 0
          }
        ]
      ]
    },
    "Simple Memory": {
      "ai_memory": [
        [
          {
            "node": "AI Agent",
            "type": "ai_memory",
            "index": 0
          },
          {
            "node": "Chat Memory Manager",
            "type": "ai_memory",
            "index": 0
          }
        ]
      ]
    },
    "When chat message received": {
      "main": [
        [
          {
            "node": "Chat Memory Manager",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Chat Memory Manager": {
      "main": [
        [
          {
            "node": "AI Agent",
            "type": "main",
            "index": 0
          }
        ]
      ]
    }
  },
  "pinData": {},
  "meta": {
    "templateCredsSetupCompleted": true,
    "instanceId": "5c6c7db12fc5c4b45387d3e63c704ebba8e8c2376e53c593366bf28f71b5a439"
  }
}

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:

Hello @Jaden_Lee This is actually not an error, it is just how n8n displays chat workflows and it confuses a lot of people at first.

When you use the Chat Trigger and click Open Chat to send a message, the execution happens in the background. The workflow editor does not always show the live data flow the same way a regular workflow does. This is normal behavior.

To confirm your workflow is actually running, instead of watching the canvas, go to the Executions tab at the top of your workflow. You will see all recent runs there with full details of what happened in each node.

There is one thing I noticed in your workflow JSON though. The model name you are using is models/gemini-3.1-flash-lite-preview and this model does not exist. The correct name is models/gemini-2.0-flash-lite or models/gemini-1.5-flash. Using a wrong model name will cause the AI Agent to silently fail which could be why you are not seeing any response coming back.

Fix the model name first, then test again by sending a chat message and checking the Executions tab instead of watching the canvas. You should see the full execution there with data passing through every node.

The system message and persona instructions you configured are completely fine and will not cause any timeout or execution issues.

@Jaden_Lee moosa’s right about the canvas display thing but check your executions tab — bet you’re seeing failures there. your model name models/gemini-3.1-flash-lite-preview doesn’t exist, that’ll silently kill the agent. swap it to gemini-2.0-flash and also ditch the Memory Manager node for now, it’s overcomplicating things. here’s a stripped down version that works:

plug in your Google Gemini creds on the model node, add your system message back into the agent options, and you should be good.

welcome to the n8n community @Jaden_Lee
I’d check whether the workflow is active/published, not only executed manually. The Chat Trigger is webhook-based, so clicking Execute Workflow in the editor can behave differently from the actual chat endpoint. I’d activate the workflow first, open the chat again, and then test from a fresh session. If it still stops there, I’d inspect the execution log for whether the Chat Memory Manager received any input items.