Welcome
Orchestrating
Orchestrate anything with HCL Universal Orchestrator.
AI agent definition
An AI agent uses a large language model (LLM) as its cognitive engine to interpret natural-language requests, break down complex goals, and make autonomous decisions.
Managing the AI agent
Course of action if the agent type is External MCP
Connecting an AI agent of type External MCP or Agentic AI Builder to an external client requires the generated server URL and an API key for authentication.

What's new
The What's new section tracks the capabilities delivered across each HCL Universal Orchestrator release, organized by version, so you can navigate directly to the features relevant to your deployment.
Getting started with HCL Universal Orchestrator
HCL Universal Orchestrator is a cloud-native workload orchestration and scheduling platform built on Kubernetes. Its microservices-based architecture enables the centralized orchestration of multi-cloud and hybrid environments from a single point of control.
Deploying and Administrating
Deploy, connect, and administer HCL Universal Orchestrator on your own infrastructure, or start immediately with a hosted SaaS environment where the service provider manages all platform operations.
Administering multitenancy
Configure and administer a HCL Universal Orchestrator multitenant environment.
Using the UI
The UI is the HCL Universal Orchestrator interface, from which you can design your scheduling environment and monitor every process.
Orchestrating
Orchestrate anything with HCL Universal Orchestrator.
- Orchestration CLI commands
  You can use the various command options in Orchestration CLI to achieve specific results. The command options are not case sensitive and you can use either uppercase or lowercase letters.
- Workflow definition
  A workflow is a set of tasks to be run, together with times, priorities, and other dependencies that define the processing order.
- Task definition
  A task definition is a unit of work that specifies an action to be performed on specific workstations in the HCL Universal Orchestrator network.
- AI agent definition
  An AI agent uses a large language model (LLM) as its cognitive engine to interpret natural-language requests, break down complex goals, and make autonomous decisions.
  - Managing the AI agent
    - Managing an AI agent using the UI
      You can create an AI agent by integrating workflows and Large Language Models (LLM) in a single platform. You can then communicate with the AI agent either using a chatbot interface in the UI or by running the AI. You can interact with the AI agent either from the AI Agents section in the UI or you can create tasks to run the agent.
    - Managing an AI agent using Orchestration CLI
      You can create an AI agent of basic agent type using Orchestration CLI. After you create an AI agent, you can run it as task or communicate with the agent using the chatbot function to get the required results.
    - Managing agent types in the AI agent
      You can create an AI agent using three different agent types: External MCP, Basic, and Agentic AI Builder. For the basic agent type, you must configure a few attributes in the values.yaml file to connect the AI agent to the AI models (LLM) of each agent type.
    - Course of action if the agent type is External MCP
      Connecting an AI agent of type External MCP or Agentic AI Builder to an external client requires the generated server URL and an API key for authentication.
  - Communicating with the AI agent
  - Exposing internal AI agents to external clients by using the Agent-to-Agent protocol
    You can expose selected internal HCL Universal Orchestrator AI agents as Agent-to-Agent (A2A) agents. The A2A protocol provides a standard framework for discovering and invoking agent clients.
- Cross dependencies definition
  Cross dependencies enable you to define dependencies between workflows running on different orchestration engines.
- Event source definition
  An event source is an item that allows HCL Universal Orchestrator to receive events from specific event plug-ins.
- Endpoint definition
  An endpoint is a reusable item that defines the connection parameters required to communicate with an external system.
- Calendar definition
  Calendars define the days in which workflows do or do not run.
- Folder definition
  You can use folders to organize your items accordingly to your business needs.
- Credentials definition
  Credential definitions are used to store user IDs and passwords, or tokens.
- Variable table definition
  Variables are values that can change according to the purpose of workflows and task templates.
- Run cycle group definition
  You can schedule workflows based on common run cycle groups.
- Workstation definition
  A workstation is a scheduling item that runs tasks.
- Resource definition
  A resource is an item that represents a quantifiable asset, measured in terms of units. You can add resources as dependencies in tasks or workflows.
- Event plug-in structure
- Managing matching criteria
  In this topic you can find a description of matching criteria.
- Parsing OCLI files in streaming
  HCL Universal Orchestrator enables you to add or replace files containing multiple item types through OCLI in a single command.
- Data exchange and transformation
  This topic explains how you can streamline your workflows using data exchange and transformation.
- Changing the default editor for Orchestration CLI
  To customize the editing experience in HCL Universal Orchestrator, you can modify the default editor by setting the visual or editor environment variables.
Orchestrating from Mobile
You can use your mobile devices to manage quickly and easily your whole HCL Universal Orchestrator environment.
Monitoring
Take your environment under control.
Using REST API
Use REST APIs to easily integrate workload scheduling capabilities with external products and solutions.
Uno AI Pilot
Welcome to the HCL Universal Orchestrator AI Pilot documentation. This resource is designed to guide you through everything you need to know about using UnO AI Pilot effectively. You will find detailed instructions, feature overviews, and helpful tips to ensure you get the best experience with UnO AI Pilot.
UnO Agentic AI Builder
Troubleshooting
Recover your environment if any issue occurs.
Notices

Course of action if the agent type is External MCP

Connecting an AI agent of type External MCP or Agentic AI Builder to an external client requires the generated server URL and an API key for authentication.

Figure 1. Flow of work for External MCP agent type

You can configure the AI agent to connect it with any external MCP server. This permits you to communicate with the HCL Universal Orchestrator resources via your chosen external source. You can access the external server through its entry points, such as a UI or a chatbot interface, which in turn can run job streams directly or invoke the necessary job streams in the HCL Universal Orchestrator environment based on the user queries. The specific interaction method or access interfaces of the external source depend on its technical specifications.

To establish the connection between the external source and the HCL Universal Orchestrator environment, you need two key components:

External MCP server URL

This URL is auto generated in the External MCP server URL field when you select the agent type as External (MCP) or Agentic AI Builder while creating an AI agent.
API key

Create an API key with your HCL Universal Orchestrator account. For more information, see steps 1 and 2 described in Managing API Keys from the UI.

Add the URL and the API key in the MCP source environment to create the connection. After the connection is established, the external MCP server can access the AI agent resources to manage the operations based on user inputs.