---
title: "Use Assistant health checks | Grafana Cloud documentation"
description: "Read the Assistant health check banner on Kubernetes detail pages to see a plain-language summary of detected issues, and open an Assistant conversation with full page context."
---

> For a curated documentation index, see [llms.txt](/llms.txt). For the complete documentation index, see [llms-full.txt](/llms-full.txt).

# Use Assistant health checks

> Note
> 
> Assistant health checks is currently in [public preview](/docs/release-life-cycle/). Grafana Labs offers limited support, and breaking changes might occur prior to the feature being made generally available.

Grafana Assistant runs a health check at the top of every Kubernetes detail page. **Assistant health checks** gives you a plain-language summary of issues that the Assistant detects, so you can catch problems while you navigate instead of scanning every panel.

The **Assistant health check** shows:

- A health indicator of the current state: **Healthy** or **Unhealthy**
- A one-sentence summary of what the Assistant detected

[**Assistant health check** notice on a workload detail page](/media/docs/grafana-cloud/k8s/asst-health-check.png)

## Investigate with the **Assistant health check**

To investigate a finding, click the chat icon on the far right side. The Grafana Assistant panel opens and explains what is happening and any next steps. You can ask follow-up questions of Grafana Assistant without re-explaining what you’re looking at.

## Refresh the check

Click the refresh icon to re-run the health check immediately. Use this to confirm a fix worked, or to get an updated status without waiting for the next page refresh.

## Manage preference for health checks

Click the **Manage assistant preferences** link to open the **Manage app** tab. Here you can control how the health check behaves within Kubernetes Monitoring.

[**Manage app** tab where health check preferences can be set](/media/docs/grafana-cloud/k8s/config-health-check.png)

The **Auto-run assistant health checks** toggle in **Configuration** > **Manage app** controls whether the health check runs automatically on every Kubernetes detail page.

- When the toggle is on, the **Assistant health check** executes Prometheus and Loki queries in the background on every Cluster, namespace, workload, Pod, container, and Node detail page, then displays the banner.
- When the toggle is off, the check doesn’t run automatically. You can still refresh it manually or open Grafana Assistant from a panel to ask about a page.

You may want to turn off the auto-run health check to reduce background query load and Assistant usage, or if you prefer to control when a health check is run.

The preference is saved in your browser’s local storage, so it’s a per-user, per-browser preference. This doesn’t affect teammates or other browsers.