Agent
The Agent can be used to generate Models, Instances, and Inputs.
Setup
Before using the Agent, configure an AI Connection with at least one Input.
See Model Suggestions for model recommendations and context window requirements.
Using the Agent
The Agent can propose new Models, Inputs, and Instances.
Proposed changes will only be applied after clicking the “Import”/“Import All” button. The agent does not have the ability to create/modify configuration without approval using these buttons.
The Agent can view most existing configuration, but it cannot modify existing configuration. It can only propose new elements.
Reviewing Proposals
Click on the “Preview” button to view configuration JSON for the proposed change.
Capabilities
The Agent has access to most of the product, but some information may only be available to specific sub-routines.
Generally speaking, the agent has access to information about:
- Pipelines
- Connections
- Inputs
- Models
- Instances
- Namespaces
When performing a specific action, the agent will enter a sub-routine where it may have access to additional information.
Special Sub-routine Capabilities
Browsing a Connection (Only supported for OPCUA)
- The ability to browse a connection to discover sources
- Full read access to the connection (this is not limited to existing inputs)
Creating Instances
- Read access for all existing inputs and instances
- Access to JSON Schemas for all models
Permissions
- Limited by the permissions of the user using the Agent
- Does not have Write access
- Cannot create/modify configuration without approval
Model Suggestions
Recommended models for best performance:
gpt-5.1or higherclaude-sonnet-4.5or higher
Gemini models are fully supported but have shown less consistent results in testing.
Requirements
To use the agent effectively, the model must:
- Have a sufficient context window
- Support tool calls
Context Window 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