Create and configure an agent base

Introduction

An agent-base is a configuration entity in ATRIA that represents an implementation code for an agent.

Agents base are referenced by agents to deploy the configuration of the agent.

Guidelines to configure an agent-base

1. Create a new agent-base

• Build the agent-base for your use case (json file), using the available agent base fields.

• When the agent-base json file is generated, execute this command to include it:

  curl --location --request POST 'https://svc-<env>.auracognitive.com/aura-services/v2/configuration/agents-base/' \
    --header 'Content-Type: application/json' \
    --header 'Accept: application/json' \
    --header 'Authorization: APIKEY XXX' \
    --data-raw '<NEW AGENT BASE JSON>'
  

1.1. Modify/update an agent-base

If once created, certain modifications are required, follow these instructions:

• Make the required changes in the agent-base json file using the available agent base fields.

• When the agent-base is modified, execute this command to update it:

  curl --location --request PUT 'https://svc-<env>.auracognitive.com/aura-services/v2/configuration/agents-base/<agentBaseId>' \
    --header 'Content-Type: application/json' \
    --header 'Authorization: APIKEY XXX' \
    --data '<AGENT BASE JSON WITH MODIFICATIONS>'
  

1.2. Delete an agent-base

• Execute the following command:

  curl --location --request DELETE 'https://svc-<env>.auracognitive.com/aura-services/v2/configuration/agents-base/<agentBaseId>' \
    --header 'Accept: application/json' \
    --header 'Authorization: APIKEY XXX'
  

2. Include the agent-base in the agent

Create the agents configuration, indicating de agentBase.name, as explained in the document guidelines for the configuration of an agent

Agent base fields

The fields for the characterization of an agent-base are summarized below, as defined in the API swagger Aura Configuration ATRIA Agents Base:

Metadata (metadata)

Example: agent-base configuration

{
  "id": "cd2b534c-16c3-4d89-a87e-ec45d3939232",
  "name": "agent-base-test",
  "description": "An AI agent built with langgraph that provides personalized recommendations about devices by querying and analyzing data stored in a MongoDB database.",
  "language": "python",
  "version": "1.0.0"
}

Note:

• The id, name, and language fields are mandatory.
• The language must be python.