Docs

Observability Platform

Using intelligent O&M assistant

This topic describes how to use the intelligent O&M assistant.

Prerequisites

All conversation content is generated by AI models and does not represent the opinions of the platform. Use the intelligent O&M assistant properly after acknowledging and agreeing to this statement. For legal agreement details, see AI Service Terms and Disclaimer.

Limits

Currently, four observability products including AI application monitoring, Server-side Monitoring, VMP, and Cloud Monitor have been successfully integrated with the intelligent O&M assistant. RUM and CloudDetect do not yet support access to the intelligent O&M assistant.

The intelligent O&M assistant only provides DevOps services including observability data queries, basic product analysis, and cross-service comprehensive troubleshooting. It does not offer services beyond intelligent O&M such as creating resources, bypassing system instructions, or outputting sensitive data. For more details, see Core capabilities.

Procedure

In the left navigation pane, select Intelligent O&M assistant to enter the intelligent O&M assistant interactive page.

Select an existing session or click "Start Task" to create a new session. Then click the drop-down button in the lower-left corner of the chat box and select an agent.

tip

VeOps Agent is a system preset agent that covers data sources, including Server-side Monitoring, VMP, Cloud Monitor, TLS, and VKE. It provides cross-source data query and troubleshooting capabilities.

To configure other agents, go to the "Expert Agent" page. For more information, see Custom expert agents.

Describe your issues, alert symptoms, or troubleshooting targets in natural language in the chat box. The intelligent O&M assistant will call tools for analysis and provide detailed answers.

Usage example

Enter the sentence "List the top 3 services by current request error rate and describe the error rate trend" in natural language in the chat box. The intelligent O&M assistant will interpret your question, initiate a diagnosis, and return detailed recommendations.

Connection: The intelligent O&M assistant calls multiple tools to sequentially interpret input content, query metric categories and service information, locate the top 3 services by error rate, and finally connect to metrics related to the error rates.

Diagnosis: For the located top 3 services, it analyzes error rate data from dimensions such as latest data, overall change trends, and data point change trends, and provides diagnostic conclusions.

Recommendations: It provides preliminary post-diagnosis judgments and follow-up recommendations to guide users in resolving issues and closing the loop.

Last updated: 2026.07.31 10:26:56