Getting Started

Setting up AI Services requires three steps: connecting an LLM provider, optionally adding external MCP servers, and configuring an AI Agent.

Prerequisites

  • An active subscription to a supported LLM provider (OpenAI, Anthropic, Azure OpenAI, Google AI, Ollama, Mistral, etc.)

  • API credentials for the LLM provider

  • Role permission: AI Services set to Full

Quick Setup

Step 1: Add an LLM Integration

Navigate to Administration ‣ Integrations and click + ADD. Select an LLM provider type from the available options. Configure the API key and endpoint (provider-specific), then save. Models are automatically discovered from the provider.

See LLM Integrations for provider-specific configuration details.

Step 2: Create an AI Agent

Navigate to Tools > AI Services > Agents and click + ADD. Configure:

  • Select the LLM Integration added in Step 1

  • Choose a model (e.g., GPT-4o, Claude 3 Opus)

  • Write a system prompt describing the agent’s role

  • Enable “Include Local MCP” to give the agent access to built-in Morpheus tools

  • Mark as Default to make this the account-wide assistant

See AI Agents for full configuration details.

Step 3: Use the AI chat assistant

Once an AI Agent is configured and enabled, the AI chat widget becomes available to authenticated users. Click the chat icon to start a conversation.

Example prompts:

  • “Show me all instances in the Production group”

  • “What’s the current CPU usage on my VMware cloud?”

  • “Create a new Ubuntu instance on the Dev cluster”

  • “List all failed backups from the last 24 hours”

Optional: Add External MCP Servers

To extend the AI’s capabilities beyond built-in Morpheus operations, add external MCP servers:

Navigate to Tools > AI Services > MCP Servers and click + ADD. Configure the server URL, protocol, and credentials.

See MCP Servers for details.