Skip to main content

Load Session

Loads an existing session to retrieve conversation state and configure artifact services.

Common Properties​

  • Name - The custom name of the node.
  • Color - The custom color of the node.
  • Delay Before (sec) - Waits in seconds before executing the node.
  • Delay After (sec) - Waits in seconds after executing node.
  • Continue On Error - Automation will continue regardless of any error. The default value is false.

Inputs​

  • Session ID - (string) A unique identifier for the specific conversation thread to retrieve.
  • App Name - (string) Identifies which agent application this conversation belongs to.
  • User ID - (string) Links the conversation to a particular user.

Options​

  • Session Service - Session storage backend. Options:
    • Memory - Temporary storage
    • Database - Persistent storage in a database
    • Robomotion - Managed by Robomotion API (default)
  • Database URL - (string) The database connection URL, required if Session Service is set to Database.
  • Artifact Service - Artifact storage backend. Options:
    • Memory - Temporary storage
    • GCS - Google Cloud Storage for persistent storage
    • Robomotion - Managed by Robomotion API (default)
  • GCS Bucket - (string) The name of the GCS bucket, required if Artifact Service is set to GCS.
  • GCS Service Account - (Credentials) Service account credentials for GCS access, required if Artifact Service is set to GCS.

Outputs​

  • State - (object) The state data associated with this session, containing conversation context and variables.

How It Works​

The Load Session node retrieves an existing conversation session:

  1. Input Validation - Verifies that Session ID is provided
  2. Auto-Detection - Detects if running in Robomotion environment for auto-configuration
  3. Service Selection - Connects to the appropriate session service (Memory, Database, or Robomotion)
  4. Session Retrieval - Loads the session using the Session ID, App Name, and User ID
  5. Artifact Configuration - Sets up the artifact service for the loaded session
  6. State Output - Returns the session's state data

Common Use Cases​

  • Resuming Conversations - Continue a previous conversation with an agent
  • State Recovery - Retrieve conversation context and data from a previous session
  • Multi-Turn Dialogues - Load session state for ongoing back-and-forth interactions
  • User History - Access a user's previous interactions and preferences
  • Workflow Continuation - Resume a workflow from where it left off
  • Session Migration - Load sessions created in different workflows or environments

Error Handling​

The node will return errors in the following cases:

  • Missing required fields:
    • "Session ID is required"
    • "App Name is required" (for Memory and Database services)
    • "User ID is required" (for Memory and Database services)
    • "Database URL is required when using database session service"
    • "GCS bucket is required when using GCS artifact service"
    • "GCS service account is required when using GCS artifact service"
  • Session not found - The specified session does not exist
  • Service unavailable - Cannot connect to the storage backend
  • Authentication failure - Invalid credentials for database or GCS
  • Robomotion context error - "Robomotion Session Service can only work with Agents and Application robots"
  • Permission errors - Insufficient permissions to access the session

Usage Notes​

  • Session ID Required - Unlike Create Session, Load Session always requires a Session ID
  • Service Consistency - Load the session using the same service type it was created with
  • Auto-Detection - When using Robomotion services, App Name and User ID may be auto-detected from context
  • State Availability - The loaded state contains all variables and context from the session's history
  • Artifact Access - After loading, you can access artifacts stored in the session using artifact nodes
  • Service Configuration - Both session and artifact services must be configured to match the session's storage
  • GCS Credentials - For GCS artifact service, provide valid service account credentials

Service Compatibility​

Ensure you load sessions with the correct service configuration:

  • Memory sessions - Use Session Service: Memory
  • Database sessions - Use Session Service: Database with the same Database URL
  • Robomotion sessions - Use Session Service: Robomotion

Example Usage​

Loading a session from Memory:

  • Session ID: "session-abc123"
  • App Name: "customer-support"
  • User ID: "user-12345"
  • Session Service: Memory
  • Artifact Service: Memory

Loading a session from Robomotion:

  • Session ID: "session-xyz789"
  • App Name: "sales-assistant" (or auto-detected)
  • User ID: "user-67890" (or auto-detected)
  • Session Service: Robomotion
  • Artifact Service: Robomotion

Loading a session from Database with GCS artifacts:

  • Session ID: "session-def456"
  • App Name: "data-analyzer"
  • User ID: "user-11111"
  • Session Service: Database
  • Database URL: "postgresql://user:pass@localhost:5432/agents"
  • Artifact Service: GCS
  • GCS Bucket: "my-artifacts-bucket"
  • GCS Service Account: (credentials from vault)

Best Practices​

  • Session ID Persistence - Store Session IDs in a database or user context to retrieve conversations later
  • Error Handling - Enable Continue On Error if the session might not exist
  • Service Matching - Always use the same service configuration as when the session was created
  • State Validation - Check that the loaded state contains expected keys before using it
  • Artifact Cleanup - Consider cleaning up old artifacts when loading sessions to manage storage