Skip to main content

​
Overview

The Grafana Monitoring Agent is an AI-powered operational assistant that helps users investigate and diagnose alerts generated by Grafana. It retrieves monitoring data from Grafana, Prometheus, and Kubernetes to provide contextual insights and assist with identifying the root cause of infrastructure and application issues. Instead of manually navigating dashboards and querying monitoring systems, users can interact with the Grafana Monitoring Agent to retrieve alerts, execute PromQL queries, inspect metrics, and analyze connector logs. The Grafana Monitoring Agent invokes specialized tools to gather monitoring information directly from the Condense platform and its integrated observability services.

​
Supported Operations

The Grafana Monitoring Agent can assist with:
  • Diagnosing Grafana alerts
  • Retrieving connector logs
  • Executing instant PromQL queries
  • Executing range-based PromQL queries
  • Listing available Prometheus metrics
  • Retrieving Prometheus metric metadata
  • Viewing Prometheus scrape targets

​
Available Tools

​
Typical Use Cases

The Grafana Monitoring Agent can be used to:
  • Investigate active Grafana alerts.
  • Diagnose infrastructure issues using monitoring data.
  • Execute PromQL queries without directly accessing Prometheus.
  • Inspect historical metric trends.
  • Review connector logs during incidents.
  • Validate Prometheus scrape targets.
  • Analyze infrastructure metrics during troubleshooting.
  • Assist with operational root cause analysis.

​
How It Works

When a monitoring request is submitted, the Grafana Monitoring Agent automatically determines the appropriate tools required to fulfill the request. Depending on the task, the agent may:
  • Retrieve the latest Grafana alerts.
  • Query Prometheus metrics.
  • Analyze historical metric data.
  • Retrieve connector logs from Kubernetes.
  • Correlate monitoring information across multiple platform components.
This provides users with contextual operational insights without manually accessing multiple monitoring systems.

​
Best Practices

  • Investigate alerts as soon as they are generated.
  • Use PromQL queries to validate alert conditions.
  • Review connector logs alongside infrastructure metrics.
  • Monitor scrape target health to ensure accurate metric collection.
  • Combine Grafana dashboards with AI Agent insights for faster troubleshooting.

​
Related Documentation

  • AI Agents Overview
  • Getting Started
  • Kafka Agent
  • Kubernetes Agent
  • Platform Monitoring

​
Frequently Asked Questions (FAQs)

​
What does the Grafana Monitoring Agent do?

The Grafana Monitoring Agent retrieves Grafana alerts, queries Prometheus metrics, analyzes monitoring data, and assists with diagnosing infrastructure and application issues.

​
Can the Grafana Monitoring Agent execute PromQL queries?

Yes. The agent supports both instant and range-based PromQL queries through the integrated Prometheus service.

​
Can I retrieve connector logs?

Yes. The GetConnectorLogs tool allows you to retrieve logs for deployed connectors by providing the Workspace ID, Connector ID, and the required log duration.

​
Does the agent perform root cause analysis?

The GetAlert tool retrieves the latest Grafana alert and provides contextual diagnostic information to help identify the likely cause of the alert.

​
Does the Grafana Monitoring Agent modify monitoring configurations?

No. The current release provides read-only monitoring and diagnostic capabilities. It does not modify Grafana, Prometheus, or Kubernetes configurations.

​
Does the Grafana Monitoring Agent respect Condense permissions?

Yes. The Grafana Monitoring Agent operates within the Condense Role-Based Access Control (RBAC) model and only retrieves information that the authenticated user is authorized to access.