AI · 04 — Extract structured data

Free-text email in → clean fields out. The ◈-wired Extract node pulls orderId / customer / urgent from the message so downstream nodes get real columns, not prose. Swap the Inject text for any email.

What’s inside

4 nodes — every type links to its full reference page.

StepNodeType
Email Inject input.inject
Extract fields AI Extract ai.extract
Ollama (local) AI Model ai.model
Fields Console utility.console

Import it

  1. Studio → Import — paste the JSON below (or the URL /docs/templates/ai-04-extract-structured-data.json).
  2. Or ask the AI Copilot with the JSON pasted after the phrase:
    import this workflow json into a new workflow named "AI · 04 — Extract structured data"
  3. Any credentials the nodes need are ${credential.…} references — add them once in Admin → Credentials and the template picks them up. Nothing secret ships in a template.
The workflow document (flowdrome.workflow.v1, 4 node types)
{
  "schemaVersion": "flowdrome.workflow.v1",
  "id": "",
  "name": "AI · 04 — Extract structured data",
  "version": "0",
  "status": "draft",
  "nodes": [
    {
      "id": "seed",
      "type": "input.inject",
      "label": "Email",
      "category": "trigger",
      "config": {
        "contentType": "application/json",
        "payload": {
          "text": "Hi, this is Ada Lovelace. Order 8412 arrived damaged — please rush a replacement, we need it for Friday's launch!"
        }
      },
      "position": {
        "x": 0,
        "y": 0
      }
    },
    {
      "id": "pull",
      "type": "ai.extract",
      "label": "Extract fields",
      "category": "ai",
      "config": {
        "provider": "ollama",
        "baseUrl": "",
        "apiKey": "",
        "model": "qwen2.5:1.5b",
        "systemPrompt": "",
        "inputField": "text",
        "attributes": [
          {
            "name": "orderId",
            "description": "the order number",
            "type": "string",
            "required": "yes"
          },
          {
            "name": "customer",
            "description": "the customer's name",
            "type": "string",
            "required": "no"
          },
          {
            "name": "urgent",
            "description": "whether they need it urgently",
            "type": "boolean",
            "required": "no"
          }
        ],
        "timeoutMs": 120000
      },
      "position": {
        "x": 530,
        "y": 0
      },
      "attachments": {
        "model": "llm"
      }
    },
    {
      "id": "llm",
      "type": "ai.model",
      "label": "Ollama (local)",
      "category": "ai",
      "config": {
        "provider": "ollama",
        "baseUrl": "",
        "apiKey": "",
        "model": "qwen2.5:1.5b",
        "models": [],
        "temperature": 0.2,
        "maxTokens": 512
      },
      "position": {
        "x": 530,
        "y": 222
      }
    },
    {
      "id": "out",
      "type": "utility.console",
      "label": "Fields",
      "category": "utility",
      "config": {
        "message": "order={{ $json.orderId }} customer={{ $json.customer }} urgent={{ $json.urgent }}"
      },
      "position": {
        "x": 1060,
        "y": 0
      }
    }
  ],
  "edges": [
    {
      "id": "e0",
      "sourceNodeId": "seed",
      "sourcePort": "output",
      "targetNodeId": "pull",
      "targetPort": "input"
    },
    {
      "id": "e1",
      "sourceNodeId": "pull",
      "sourcePort": "output",
      "targetNodeId": "out",
      "targetPort": "input"
    }
  ],
  "triggers": [
    {
      "id": "t0",
      "nodeId": "seed"
    }
  ],
  "meta": {
    "snippets": [],
    "source": "template-gallery",
    "description": "Free-text email in → clean fields out. The ◈-wired Extract node pulls orderId / customer / urgent from the message so downstream nodes get real columns, not prose. Swap the Inject text for any email."
  }
}

Generated from the verified demo corpus — this exact document seeds and runs on a fresh Flowdrome. Raw JSON: ai-04-extract-structured-data.json.