AI Chat

更新时间:
复制 MD 格式

AI Chat lets you explore observability data, detect anomalies, diagnose faults, and perform root cause analysis by conversing with an AI-powered O&M assistant in natural language.

Overview

AI Chat is the natural language interface of STAROps (Intelligent O&M Platform). Describe what you need in plain language, and the AI assistant retrieves data, runs analyses, and presents results — no complex queries or dashboard navigation required.

Core capabilities

AI Chat supports the following capabilities:

  • Multi-turn conversations: Context is maintained across turns, so you can drill down into issues without repeating background information.

  • Tool-based task execution: Built-in tools query metrics, analyze logs, and diagnose issues on your behalf.

  • Rich result presentation: Results appear as text explanations, data tables, metric trend charts, or topology diagrams based on the query type.

Supported scenarios

Scenario

Description

Example prompt

Data exploration

Query observability data using natural language.

"Show the top 5 hosts by CPU usage over the past 24 hours."

Anomaly detection

Identify metric anomalies and alert events.

"What alerts were triggered in the last hour?"

Fault diagnosis

Narrow down issues through multi-turn conversations.

"Analyze the response latency increase for the payment service."

Root cause analysis

Perform correlation analysis using the Unified Observability Model (UModel) topology.

"What is the root cause of this alert?"

Entry points

You can open AI Chat from the following entry points:

Entry point

Description

CloudMonitor (CMS) console

Click Intelligent O&M Assistant in the CMS sidebar. Supports @entity references.

Log Service (SLS) console

Select a project and Logstore in SLS, then open the conversation panel. The current SLS context is automatically injected.