Skip to content

Latest commit

 

History

History
executable file
·
47 lines (37 loc) · 1.88 KB

File metadata and controls

executable file
·
47 lines (37 loc) · 1.88 KB
subcategory Agent Bricks

databricks_supervisor_agent Resource

Public Beta

API Documentation

Example Usage

Arguments

The following arguments are supported:

  • display_name (string, required) - The display name of the Supervisor Agent, unique at workspace level
  • description (string, optional) - Description of what this agent can do (user-facing)
  • instructions (string, optional) - Optional natural-language instructions for the supervisor agent
  • provider_config (ProviderConfig, optional) - Configure the provider for management through account provider.

ProviderConfig

  • workspace_id (string,optional) - Workspace ID which the resource belongs to. This workspace must be part of the account which the provider is configured with.

Attributes

In addition to the above arguments, the following attributes are exported:

  • create_time (string) - Creation timestamp
  • creator (string) - The creator of the Supervisor Agent
  • endpoint_name (string) - The name of the supervisor agent's serving endpoint
  • experiment_id (string) - The MLflow experiment ID
  • id (string, deprecated) - Deprecated: Use supervisor_agent_id instead
  • name (string) - The resource name of the SupervisorAgent. Format: supervisor-agents/{supervisor_agent_id}
  • supervisor_agent_id (string) - The universally unique identifier (UUID) of the Supervisor Agent

Import

As of Terraform v1.5, resources can be imported through configuration.

import {
  id = "name"
  to = databricks_supervisor_agent.this
}

If you are using an older version of Terraform, import the resource using the terraform import command as follows:

terraform import databricks_supervisor_agent.this "name"