- CPU
- MEMORY
- CLUSTER
- HUMAN
Short answer
How can I get system metrics using my AI?
Your agent uses the query_data tool to execute DQL statements against Guance Cloud. This pulls CPU, memory, and cluster-specific data straight to your chat. You'll see exactly how your infrastructure is performing through the following breakdown.
The data you'll get.
Direct access to telemetry.
The agent handles the query syntax so you can focus on the numbers.
CPU
Processor stats
The agent pulls usage percentages and load averages. You'll see exactly how much processing power your nodes are using.
MEMORY
RAM utilization
Get clear views of memory consumption and available capacity. This helps you spot leaks or exhaustion quickly.
CLUSTER
Node health
Query specific cluster metrics to check the status of your distributed systems. You'll get the health data for individual nodes.
HUMAN
Manual fixes
The AI only retrieves data. You still have to log in to Guance Cloud to change settings or restart services.
The workflow
What your AI does when you ask for data.
The agent acts as the bridge between your question and the Guance Cloud database.
Interpret request
Your agent looks at your question and determines which metrics and time ranges you need.
query_dataBuild DQL
The agent converts your request into a valid DQL statement for Guance Cloud.
query_dataFetch data
The agent executes the query to pull the raw telemetry from your environment.
query_dataPresent results
The agent cleans up the raw output and presents the metrics in a readable format.
query_data
Try it
Copy these to start.
Paste these into your chat to test the connection.
Starting points
Replace the cluster names and metric types with your own specific identifiers.
Guance Cloud / 观测云 Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_AANzdKMqG2pZyVDyBezhkEvdo4gPXGC9YiQz8RcU/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Guance Cloud / 观测云 capabilities are ready to use.
{
"mcpServers": {
"guance-cloud-mcp": {
"url": "https://edge.vinkius.com/vk_preview_AANzdKMqG2pZyVDyBezhkEvdo4gPXGC9YiQz8RcU/mcp"
}
}
}Show me the average CPU usage for the production-cluster over the last 15 minutes.
What is the memory usage for the node named worker-01?
Get the disk utilization metrics for my database cluster.
List the top 3 nodes with the highest CPU load right now.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Guance Cloud once, then ask.
Just follow the link to authorize the connection. Your credentials stay encrypted, and you can start querying immediately.
Connect Guance Cloud to your AIFAQ
How the querying behaves
- 01
Can the AI change my cluster settings?
No. The agent only has permission to read data. It cannot modify your infrastructure.
- 02
Can the AI delete my logs or metrics?
No. The tool is strictly for querying data, not for deleting or altering it.
- 03
Do I need to write DQL myself?
No. Your agent writes the DQL for you based on your plain English questions.
- 04
Can I query specific time ranges?
Yes. You can ask for data from the last hour, the last day, or any other specific window.
- 05
What happens if the query fails?
Your agent will report the error returned by Guance Cloud so you can refine your request.
More questions about Guance Cloud? See everything the Guance Cloud Connector can do
Connect Guance Cloud / 观测云 to Claude, Cursor, ChatGPT & more
