Self Service AI agents

Next-generation platformThe AI agents are part of a powerful suite of IBM® watsonx™-powered AI agents. The AI agents deliver proactive support with real-time insights to administrators. Using advanced natural language processing, these agents interpret prompts and queries to run tasks, and can make it easier to manage Self ServiceIBM Sterling® Order Management System SaaS environments.

Note: AI agent trial program now available. Access and explore the latest Self Service AI agents for two months in a development or preproduction environment through our AI agent trial program. We also invite you to share your insights, and partner with us to shape the future of AI agent-powered support. Request access to the AI agent trials through IBM Support. For more information, see Requesting access to the AI agent trials.
Note: The AI agent trial program is designed for IBM Sterling Order Management System SaaS customers today, with plans to extend support to on-premises environments in the future.

New to Self Service AI agents? Watch these videos to learn about the AI agents that you can use for smarter system management and recommended insights.

System administration utility agents

The following System Administration Utility agents are part of this suite.
Server Management Agent
The Server Management Agent streamlines server administration tasks by providing intuitive tools to create, update, start, stop, and restart the server.

Sample requests and prompts
  • Show all servers that have a higher than the default for running instances.
  • List my servers sorted by running instances.
Schedule Management Agent
The Schedule Management Agent handles all scheduling-related tasks, allowing you to view, create, and update schedules efficiently. It provides smart scheduling recommendations that are based on existing schedules and upcoming events to avoid conflicts and optimize resource utilization. The agent analyzes historical scheduling patterns to suggest optimal timing for new tasks.

Sample requests and prompts
  • When is the best time to schedule a new server?
  • Recommend when we should set up server schedules (when we don't have peak loads).
Event Calendar Agent
The Event Calendar Agent manages environment upgrades and maintenance events across the system. It provides a centralized view of all planned activities, helping teams coordinate their work and avoid scheduling conflicts.

Sample requests and prompts
  • When was my most recent calendar event?
  • Show all maintenance events for next week.
  • When was the last time we uploaded a customization extension?
Queue Management Agent
The Queue Management Agent provides comprehensive queue administration capabilities, including creating, updating, viewing, and configuring queues.

Sample requests and prompts
  • Get the oldest message from the ADJUST_INVENTORY queue.
Diagnostics Agent
The Diagnostics Agent provides quick access to export applications, agents, integration, and IBM MQ logs.

Sample requests and prompts
  • Export app server logs for my API server.
Process Management Agent
The Process Management Agent monitors system processes to help ensure smooth operations. It provides real-time visibility into process status, identifies problematic processes, and provides a comprehensive summary on failures.

Sample requests and prompts
  • What was my last failed process?

Performance monitoring agents

The following Performance Monitoring agents are part of this suite.
Resource Optimizer Agent
The Resource Optimizer Agent intelligently recommends scaling workloads up and down based on performance KPIs to help ensure efficient resource allocation and reduce costs. It analyzes resource utilization patterns, identifies optimization opportunities, and implements the right-sizing recommendations. The agent balances performance requirements with cost considerations to achieve optimal resource efficiency across the environment.

Sample requests and prompts
  • Provide a recommendation on server resource utilization.
Observability Agent
The Observability Agent provides comprehensive monitoring and analytics across multiple system dimensions. It tracks application performance KPIs such as throughput, error rates, and latency, and platform metrics such as memory, CPU, and heap usage. It also tracks queue statistics, and database health. The agent delivers in-depth performance analysis, pinpoints bottlenecks, and offers actionable insights to facilitate optimization and proactive issue resolution before any negative impact happens to business operations. Agent conclusions are limited to data that is available in monitoring dashboards. Data is limited in QA and development environments.

Sample requests and prompts
  • Show me the agent servers that are least used by the CPU.
  • How is the performance of my API server?
  • Do I have any database lock waits?
Health Check Agent
The Health Check Agent provides a holistic assessment of the application environment. It evaluates service availability, integration health, queue status, and overall system readiness. The agent returns a structured summary of findings and recommended next steps, enabling administrators to quickly assess the health of their environment. Agent conclusions are limited to data that is available in monitoring dashboards.

Sample requests and prompts
  • Perform a health check.
IBM Sterling™ Intelligent Promising (SIP) Agent
Monitor SIP performance metrics, such as events published, response times, and API call throughput.

Sample requests and prompts
  • How is the reserve inventory API performing?
Business Performance Metrics Agent
Track business-critical metrics, including order lines that are created, scheduled, released, and failed and other KPIs.

Sample requests and prompts
  • How many order lines have been processed in the last week?

Insight and guidance agents

The following Insight and Guidance agent is part of this suite.
Search-Action Agent
The Search‑Action Agent finds answers instantly. It analyzes documentation and configuration details and surfaces clear guidance and recommended next steps. It helps you troubleshoot faster without having to sift through detailed documentation or logs. The agent is designed to accelerate problem‑solving by turning complex questions into direct, actionable insights.

Sample requests and prompts
  • How do I improve JMS PUT performance?
  • What is the default heap memory allocation for servers if we don't specify xms/xmx?

Key capabilities

The Self Service AI agents offer the following key capabilities.

Intelligent understanding
The AI agents use large language models to comprehend and respond to natural language inputs and text commands.
Contextual awareness
The AI agents understand the conversation context, retrieve relevant information from various systems, databases, and APIs, and provide responses and recommendations.
User‑controlled write operations
All write operations that are initiated by an AI agent require explicit user confirmation. When a requested action modifies system data or configuration, the agent displays a confirmation prompt that summarizes the change. The operation proceeds only after you approve it.
Feedback submission
You can submit written feedback directly through the agent interface. Feedback is routed to the development team to improve accuracy, usability, and future enhancements.
Streaming response in real-time
Streaming response in real-time provides a way for you to view the agent’s step by step thought process as it interprets and responds to a request. This feature increases clarity and helps you understand how conclusions are reached.
Response control
You can stop an agent response at any time. When a response is stopped, processing halts immediately and the system does not perform any write operations.

The Self Service AI agents are designed to simplify system interaction and provide proactive support for System Administrators. With its ability to understand natural language prompts, questions, and requests, the Self Service AI agent can assist with a wide range of tasks, from system administration to monitoring and analytics.

Note: The AI agents can access data only from the environment where the conversation takes place.

The Self Service AI agents can help with the following key areas.

System administration
The Self Service AI agents streamline system administration by providing intuitive tools for server management, smart scheduling, and resource optimization. Specifically, they can assist with the following system administration tasks.
  • Server management: Create, update, start, stop, restart, and schedule servers.
  • Schedule management: View, create, update, and delete schedules, with smart scheduling recommendations based on existing schedules and upcoming events.
  • Event calendar management: Manage environment upgrades and maintenance events.
  • Queue management: Create, update, view, and configure queues, including browsing pending tasks, prioritizing tasks, and optimizing queue performance.
  • Resource optimization: Scale workloads up and down based on performance KPIs to help ensure efficient resource allocation and reduce costs.
  • Process management: Manage and view process status to help ensure smooth operation.
Monitoring and analytics
The Self Service AI agents provide in-depth performance analysis, pinpointing bottlenecks and offering actionable insights to facilitate optimization and proactive issue resolution. Specifically, the Self Service AI agents can help with the following monitoring and analytics tasks.
  • Application performance monitoring: Track key performance indicators (KPIs) such as throughput, error rates, latency, and external service performance.
  • Platform performance monitoring: Monitor system resources, including memory, CPU usage, heap, garbage collection, and thread utilization.
  • Queue performance monitoring: Analyze queue statistics, including connection counts and queue depths.
  • Database performance monitoring: Provide insights into database health, including disk size, top tables, maintenance, long-running queries, and lock-wait status.
  • SIP performance monitoring: Monitor SIP performance metrics, such as events published, response times, and API call throughput.
  • Business performance monitoring: Track business-critical metrics, including order lines that are created, scheduled, released, and failed, inventory activity, and other KPIs.
  • Performance recommendations: Provide documentation-based recommendations when you ask performance-related questions.
  • Health checks: Run comprehensive environment health assessments by using the Health Check Agent.

Requesting access to the AI agent trials

To participate in the AI agent trials, request access through IBM Support.
  1. Click this link to open a Support case.
  2. For Type of support, select Product support.
  3. For Case title, enter Request access to AI agent trial program.
  4. For Product, enter Order Management on Cloud.
  5. For Severity, select 4 - minimal impact.
  6. In the Case description, include the environment details about where you want to enable the AI agents (only available in lower environments, not production environments).
  7. Complete the remaining items in the case form and click Submit case.

Finding AI documentation

For more information about AI capabilities and links to other AI-related documentation, see AI agent capabilities in IBM Sterling Order Management System.