> ## Documentation Index
> Fetch the complete documentation index at: https://docs.eachlabs.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Workflow Builder

> Build or update multi-step AI workflows using natural language.

## Endpoint

```
POST https://eachsense-agent.core.eachlabs.run/workflow
```

## Request Body

| Parameter     | Type    | Required | Description                                      |
| ------------- | ------- | -------- | ------------------------------------------------ |
| `message`     | string  | Yes      | Workflow description or modification instruction |
| `workflow_id` | string  | No       | Existing workflow to update                      |
| `version_id`  | string  | No       | Version to update                                |
| `stream`      | boolean | No       | Enable SSE streaming (default: `true`)           |
| `session_id`  | string  | No       | Session ID                                       |

## Code Examples

### Create a New Workflow

<CodeGroup>
  ```bash cURL theme={"dark"}
  curl -X POST https://eachsense-agent.core.eachlabs.run/workflow \
    -H "Content-Type: application/json" \
    -H "X-API-Key: YOUR_API_KEY" \
    -d '{
      "message": "Create a workflow that generates a portrait image then animates it into a 5-second video",
      "stream": true
    }'
  ```

  ```python Python theme={"dark"}
  import requests

  response = requests.post(
      "https://eachsense-agent.core.eachlabs.run/workflow",
      headers={
          "Content-Type": "application/json",
          "X-API-Key": "YOUR_API_KEY"
      },
      json={
          "message": "Create a workflow that generates a portrait then animates it",
          "stream": False
      }
  )
  print(response.json())
  ```
</CodeGroup>

### Update an Existing Workflow

```bash theme={"dark"}
curl -X POST https://eachsense-agent.core.eachlabs.run/workflow \
  -H "Content-Type: application/json" \
  -H "X-API-Key: YOUR_API_KEY" \
  -d '{
    "message": "Add a final step to upscale the video to 4K",
    "workflow_id": "wf_abc123",
    "version_id": "v1",
    "stream": true
  }'
```

## Streaming Response Events

```
data: {"type":"status","message":"Creating workflow..."}

data: {"type":"workflow_created","workflow_id":"wf_abc123","version_id":"v1","input_schema":{...},"steps_count":3}

data: {"type":"text_response","content":"I've created your workflow!"}

data: {"type":"complete","status":"ok","workflow_id":"wf_abc123"}

data: [DONE]
```

### Update Events

```
data: {"type":"workflow_fetched","workflow_name":"portrait-to-video","existing_steps":3}

data: {"type":"workflow_built","steps_count":4,"definition":{...}}

data: {"type":"workflow_updated","success":true,"workflow_id":"wf_abc123","version_id":"v2"}

data: {"type":"complete","status":"ok"}

data: [DONE]
```

## Non-Streaming Response

```json theme={"dark"}
{
  "success": true,
  "workflow_id": "wf_abc123",
  "version_id": "v1",
  "definition": {
    "steps": [...],
    "input_schema": {...}
  },
  "message": "Workflow created successfully"
}
```