Custom Flows

Build sophisticated verification experiences with FaceSign's node-based flow system. Create conditional verification paths, implement complex business logic, and design user experiences that adapt to different scenarios.

Overview

While the verification modules provide a simple way to configure common verification patterns, custom flows give you complete control over the verification process. Flows are built using a node-graph system where:

• Nodes define verification steps and actions
• Edges define the connections between nodes
• Outcomes determine which path to take based on results

Flow Structure

A flow consists of two main components:

Nodes Array

Each node represents a step in your verification process. Nodes have specific types and configurations that determine what happens at that step.

Edges Array

Edges connect nodes together, defining the possible paths through your flow based on outcomes and conditions.

Available Node Types

FaceSign supports 11 different node types for building flows:

Basic Flow Example

Here's a simple flow that performs liveness detection followed by document scanning:

The corresponding JSON structure for this flow:

{
  "nodes": [
    { "id": "start", "type": "start" },
    { 
      "id": "liveness", 
      "type": "liveness_detection",
      "outcomes": {
        "livenessDetected": "document",
        "deepfakeDetected": "end",
        "noFace": "end"
      }
    },
    {
      "id": "document",
      "type": "document_scan",
      "scanningMode": "SINGLE_SIDE",
      "allowedDocumentTypes": ["MRTD_TYPE_PASSPORT"],
      "outcomes": {
        "scanSuccess": "end",
        "userCancelled": "end",
        "scanTimeout": "end"
      }
    },
    { "id": "end", "type": "end" }
  ],
  "edges": [
    { "id": "e1", "source": "start", "target": "liveness" },
    { "id": "e2", "source": "liveness", "target": "document" },
    { "id": "e3", "source": "liveness", "target": "end" },
    { "id": "e4", "source": "document", "target": "end" }
  ]
}

Using Custom Flows

Include your custom flow in the session creation request:

const session = await client.sessions.create({
  clientReferenceId: 'user-123',
  metadata: { userId: '123' },
  flow: myCustomFlow,
  // Don't include modules when using custom flows
})

Best Practices

• Always start with a START node and end with an END node
• Ensure all possible outcome paths are connected
• Test flows thoroughly with different scenarios
• Use meaningful node IDs for easier debugging
• Consider user experience when designing conditional paths