Skip to main content

​
What are Flows?

Flows are containers that group related steps together and provide:

Shared Memory

Each flow maintains its own context that persists across steps within the flow

Component Management

Flows can have dedicated components like memory systems, specialized tools, or custom handlers

Context Transfer

When transitioning between flows, context is intelligently summarized and passed along

Entry/Exit Points

Define which steps can enter or exit a flow for better control flow management

​
Flow Configuration

You can define flows in your YAML configuration: 
# Basic agent configuration
name: advanced-assistant
persona: A helpful assistant with specialized workflows
start_step_id: greeting

steps:
  - step_id: greeting
    description: Greet the user and understand their needs
    routes:
      - target: order_taking
        condition: User wants to place an order
      - target: customer_support
        condition: User needs help or support

  - step_id: order_taking
    description: Handle order details and preferences
    available_tools:
      - get_menu_items
      - add_to_cart
    routes:
      - target: order_confirmation
        condition: Order is complete

  - step_id: order_confirmation
    description: Confirm order details and process payment
    available_tools:
      - calculate_total
      - process_payment
    routes:
      - target: farewell
        condition: Order is confirmed

  - step_id: customer_support
    description: Handle customer inquiries and issues
    available_tools:
      - search_knowledge_base
      - escalate_to_human
    routes:
      - target: farewell
        condition: Issue is resolved

  - step_id: farewell
    description: Thank the user and end the conversation

# Enhanced flows configuration
flows:
  - flow_id: order_management    # Can also use 'id'
    description: "Complete order processing workflow"  # Can also use 'desc'
    enters:
      - order_taking
    exits:
      - order_confirmation
      - farewell
    components:
      memory:
        llm:
          provider: openai
          model: gpt-4o-mini
        retriever:
          method: embedding
          kwargs:
            k: 5

  - flow_id: support_workflow
    description: "Customer support and issue resolution"
    enters:
      - customer_support
    exits:
      - farewell
    components:
      memory:
        llm:
          provider: openai
          model: gpt-4o-mini
        retriever:
          method: embedding

​
Flow Memory and Context

Each flow can have its own memory system that:
Maintains conversation history and important details within the flow
Uses BM25 or other retrieval methods to find relevant information
Automatically summarizes context when exiting a flow
Passes summarized context when transitioning between flows

​
Flow Benefits

  • Architecture
  • Context
  • Scalability
  • Memory
  • Control
Organized Architecture: Keep related functionality grouped togetherFlows help you organize your agent’s capabilities into logical groups, making your codebase more maintainable and easier to understand.

​
Example: Barista Agent with Flows

The barista example demonstrates flow usage for order management: 
flows:
  - flow_id: take_coffee_order
    description: "Complete coffee ordering process"
    enters:
      - take_coffee_order
    exits:
      - finalize_order
      - end
    components:
      memory:
        llm:
          provider: openai
          model: gpt-4o-mini
        retriever:
          method: embedding
This flow ensures that all order-related context (customer preferences, cart contents, order history) is maintained throughout the ordering process and properly summarized when the order is complete.

​
Multi-Flow Applications

You can create sophisticated applications by combining multiple flows: 
flows:
  - flow_id: authentication
    description: "User login and verification"
    enters: [login]
    exits: [dashboard, main_menu]

  - flow_id: order_processing
    description: "Handle customer orders"
    enters: [order_taking]
    exits: [checkout, main_menu]

  - flow_id: customer_support
    description: "Handle customer inquiries"
    enters: [support_request]
    exits: [resolution, escalation]

​
Flow Transitions

Flows automatically handle context transfer when users move between different workflows, ensuring seamless user experience while maintaining relevant information.

​
Best Practices

Single Responsibility

Each flow should handle one specific type of workflow or business logic

Clear Entry Points

Define clear conditions for when users should enter each flow

Context Management

Keep flow contexts focused and relevant to avoid information overload

Exit Strategies

Always provide clear exit paths from each flow

​
Configuration Validation

Use the nomos validate command to check your flow configuration: 
nomos validate config.agent.yaml --verbose
This validates:
  • Flow ID uniqueness
  • Entry/exit step references
  • Component configuration
  • Field naming compatibility (supports both id/flow_id and desc/description)
Field Naming FlexibilityNOMOS supports both compact (id, desc) and descriptive (flow_id, description) field names for flows. You can use either convention consistently in your configuration.
⌘I