Agent

The Pipeline Agent uses Inputs on AI Connections to create and modify pipelines through natural language prompts. The Agent can also answer questions about pipeline configuration and other Intelligence Hub resources.

The Agent operates with the same permissions as the current user. All proposed changes require explicit approval before being applied.

Setup

Before using the Pipeline Agent, configure an AI Connection with at least one Input.

See Model Suggestions for model recommendations and context window requirements.

Modifying a Pipeline

Making Changes

  1. Open a pipeline or create a new one
  2. Navigate to the Agent tab in the right sidebar
  3. Select an AI Connection Input from the dropdown
  4. Enter a prompt describing the desired changes
  5. Click Send

The Agent analyzes the request and current configuration, then proposes changes. Tool usage and actions are shown during processing.

Review Mode

When the Agent proposes changes, the pipeline enters Review Mode. Modified, added, or removed stages are highlighted in the graph.

In Review Mode:

  • View the modified pipeline in the graph
  • Test changes using Debug mode
  • Manually edit stages if needed
  • Accept or dismiss the proposed changes

Three options are available for handling changes:

  • Dismiss - Reverts the pipeline to its state before the Agent’s changes (preserves unsaved changes made prior to using the Agent)
  • Accept - Applies the changes without saving, creating a checkpoint for further iteration
  • Accept & Save - Applies and saves the changes to the runtime

Agent results review mode example

Chat History

The Agent maintains chat history during the session. Chat history is cleared when navigating away from the pipeline or clicking the “Start Over” button.

Debugging

The Agent can troubleshoot pipeline issues by analyzing debug information and logs.

Using Debug Data

To provide debug transaction data to the Agent:

  1. Click the Bug icon to show the Debug Menu
  2. Click Run to execute a debug session

Once a debug run is active, debug transaction information is automatically included with Agent requests.

With debug data available, the Agent can:

  • Analyze event tree execution flow
  • Examine event values and metadata passed between stages
  • Identify failed data transformations
  • Suggest fixes based on execution results

Example prompts:

  • “Why is the filter stage not passing any events?”
  • “What value is the model stage receiving?”
  • “Help me fix the error in the transform stage”

Asking Questions

Beyond modifying and debugging pipelines, the Agent can answer questions about pipeline configuration and other Intelligence Hub resources.

Example prompts:

  • “How does this pipeline compare to PipelineX?”
  • “What is this Transform Stage doing?”
  • “Which model does my OPC_Machine Instance use?”
  • “How is InputA using its parameters?”

Model Suggestions

Recommended models for best performance:

  • gpt-5.1 or higher
  • claude-sonnet-4.5 or higher

Gemini models are fully supported but have shown less consistent results in testing.

Requirements

To use the agent effectively, the model must:

  1. Have a sufficient context window
  2. Support tool calls

Context Windows Requirements

The Agent requires significant context to operate effectively. Recommended context windows:

  • Minimum: 10,000 tokens
  • Recommended: 128,000+ tokens

Insufficient context leads to inconsistent results:

  • Unable to modify pipelines effectively
  • Failed JSON generation for tool calls

To verify adequate context, ask the Agent to “List all available tools and their parameters”. If the Agent accurately lists all available tools, it likely has sufficient context.

Available Tools

The following tools are available to the Agent. In the UI, tools are displayed with friendly names. Hover over tool calls to see the actual tool name.

Tool Names

Pipeline Configuration Tools:

  • get-current-pipeline - Retrieve the current (unsaved) pipeline being edited
  • update-current-pipeline - Update the current pipeline with proposed changes
  • get-pipeline - Retrieve the configuration of a specific pipeline by name
  • get-example-pipelines - Access example pipeline configurations

List Tools:

  • list-pipelines - Get all available pipeline names
  • list-connections - Get all available connections
  • list-inputs - Get all available inputs
  • list-instances - Get all available instances
  • list-namespaces - Get all available namespaces
  • list-references - Get all available references

Debug Tools:

  • get-logs - Retrieve a list of recent log entries
  • get-debug-tree(when Debug is active) - Retrieve the execution event tree
  • get-debug-tree-event (when Debug is active) - Get detailed data for a specific debug event

Schema Tools:

  • get-instance-parameters - Get parameter and output schemas for instances
  • get-input-parameters - Get parameter schemas for connection inputs
  • get-pipeline-parameters - Get parameter schemas for sub-pipelines
  • get-output-settings-schema - Get schema for output configuration