Flows

Flows are the core orchestration mechanism in MindedJS that define how your AI agent processes conversations through a series of connected nodes. A flow represents a complete run of an AI agent that handles user interactions from trigger to completion.

What are Flows?

Flows define the conversation logic and decision-making pathways for your AI agent. Each flow consists of:

• Nodes - Processing units that handle specific tasks

• Edges - Routing logic that connects nodes

• Memory - Shared state that persists throughout the flow

• Configuration - Settings that control flow behavior

name: Customer Support Flow
nodes:
  - type: trigger
    triggerType: manual
    name: Customer Request
  
  - type: promptNode
    name: Support Agent
    prompt: |
      You are a helpful customer support agent.
      Assist the customer with their inquiry.
    llmConfig:
      name: ChatOpenAI
      properties:
        model: gpt-4o

edges:
  - source: Customer Request
    target: Support Agent
    type: stepForward

Flow Components Overview

Component	Purpose	Description
Nodes	Processing	Handle specific tasks like LLM calls, tool execution
Edges	Routing	Define transitions and decision logic between nodes
Memory	State	Shared context that persists throughout the flow
Configuration	Settings	Flow-level settings and metadata

Basic Flow Structure

Every flow follows this YAML structure:

name: Flow Name                # Required: Human-readable flow identifier
nodes:                        # Required: Array of processing nodes
  - type: trigger             # Node type
    name: Node Name           # Unique node identifier
    # ... node-specific config
    
edges:                        # Required: Array of routing connections
  - source: Source Node       # Source node name
    target: Target Node       # Target node name
    type: stepForward         # Edge type
    # ... edge-specific config

Nodes in Flows

Flows orchestrate different types of nodes to handle various aspects of conversation processing. For detailed information about each node type, see the Nodes documentation.

Edges in Flows

Edges define how your flow moves between nodes based on different conditions. For comprehensive information about edge types and routing logic, see the Edges documentation.

Memory in Flows

Memory provides shared state that persists throughout flow execution, allowing nodes to access and modify data from previous steps.