Prerequisite requirements for the IBM App Connect Enterprise Agent
Ensure that you understand the prerequisite requirements before you attempt to enable the IBM® App Connect Enterprise Agent in an App Connect Dashboard instance.
The prerequisites for enabling and interacting with the Agent are as follows:
- IBM App Connect Operator 13.0.0 or later must be deployed in your cluster.
- The Operator can either be installed independently or as part of an IBM Cloud Pak for Integration deployment on Red Hat® OpenShift® or Kubernetes. For information about installing the App Connect Operator, see Installing IBM App Connect on Red Hat OpenShift or Kubernetes environments.
- An App Connect Dashboard instance at version 13.0.7.0-r1 or later is
needed to enable the Agent. You can enable the Agent in a new Dashboard instance that is being created, or
in an existing instance. For more information, see Enabling the IBM App Connect Enterprise Agent in an App Connect Dashboard instance.Note: If Keycloak is enabled to secure access to the Dashboard instance, users must be assigned the
dashboard-adminrole for the Dashboard instance. This role grants full user access with administrative privileges to the Dashboard instance and is needed for interacting with the Agent. For more information about assigning this role, see Implementing identity and access management for App Connect Designer and App Connect Dashboard instances. - The integration runtimes that the Dashboard manages, and which the Agent can query, must be at version 13.0.5.0-r1 or later.
- Access is required to a specific set of large language models (LLMs) that are hosted in an IBM
watsonx.ai Runtime service on IBM Cloud.
The Agent requires access to these LLMs to understand your queries and formulate responses. To enable this access, you need the following watsonx.ai artifacts:
- A provisioned IBM
watsonx.ai Runtime service
instance
For the supported region, see Limitations of the IBM App Connect Enterprise Agent.
- The watsonx.ai URL that is used to call watsonx.ai APIs such as LLM inferencing, embedding, training, and chatting
- The project ID of a project (that is, a collaborative workspace) that is associated with the service
- An API key for secure authorization
For information about provisioning a service instance, see Creating and managing IBM Cloud services and Signing up for IBM watsonx in the IBM Cloud documentation.
If a project has already been created for use in your service instance, you can find details about the project ID, watsonx.ai URL, and API key on the IBM watsonx Home page. The following image shows an example of the Home page with a Developer access panel that provides the watsonx.ai details that you need. If you want to start by creating a project and API key, see Creating a project in the IBM Cloud documentation.
Note: Entitlement to use IBM watsonx.ai is separate to your IBM App Connect Enterprise software entitlement. For more information, see watsonx.ai Runtime service plans. - A provisioned IBM
watsonx.ai Runtime service
instance