Flo Agentic Workflows icon

Flo Agentic Workflows

Interact with Flo Agentic Workflows
Join our community

Overview

This node, named "Flo Agentic Workflows," allows users to interact with predefined workflows hosted on a Llama Deploy server. It sends a user-defined message to a selected workflow and optionally evaluates the result returned by that workflow. This is useful for automating interactions with AI-driven or agentic workflows such as compliance assistants or incident management systems.

Common scenarios include:

  • Sending queries or commands to an AI-powered policy and compliance assistant to get guidance or check compliance.
  • Reporting incidents or requesting actions from an incident management workflow.
  • Automating conversational workflows where the input message drives the workflow logic and the output is processed further in n8n.

Practical example:
A user inputs a message describing a compliance question, selects the "Policy and Compliance Assistant" workflow, and receives a structured response that can be used downstream in the automation.

Properties

Name Meaning
Deployment Name The name of the deployment on the Llama Deploy server. Defaults to "StateMachines".
Workflow Name Selects which workflow to send the message to. Options:
- Policy and Compliance Assistant
- Incident Management
Evaluate Result Boolean flag indicating whether to evaluate the result of the workflow ("True" or "False").
User Message The message text sent to the Llama Deploy server. Can use expressions referencing input data (e.g., {{ $json.chatInput }}). Required field.

Output

The node outputs an array of JSON objects, each corresponding to one input item processed. Each output JSON contains:

  • response: The main content returned by the workflow after processing the input message. This is typically a string or structured data parsed from the workflow's result.
  • In case of failure, the output JSON includes:
    • success: false
    • error: A brief error message ("Failed to send message")
    • details: Detailed error information including the original workflow name and input parameters.

The node does not output binary data.

Dependencies

  • Requires connection to a Llama Deploy server via an API key credential.
  • Needs configuration of API server URL and control plane URL through credentials.
  • The node uses an internal client (LlamaDeployClient) to manage sessions and run workflows on the server.
  • Timeout for workflow execution can be configured via credentials (default 120 seconds).

Troubleshooting

  • Common issues:

    • Failure to connect to the Llama Deploy server due to incorrect API URLs or missing/invalid API key credential.
    • Workflow execution timeout if the server takes too long to respond.
    • Malformed or unexpected responses from the workflow causing JSON parsing errors.

  • Error messages:

    • "Failed to send message": Indicates the node could not successfully invoke the workflow. Check network connectivity, API credentials, and workflow availability.
    • "Failed to clean up session" (logged to console): Indicates an issue deleting the session after execution; usually non-critical but may indicate resource leaks on the server side.

  • Resolutions:

    • Verify API credentials and URLs are correctly set.
    • Ensure the selected workflow exists and is active on the Llama Deploy server.
    • Increase timeout if workflows take longer to process.
    • Inspect detailed error messages in the output for clues.

Links and References

Discussion