How to Pre-Configure Grafana Assistant for Instant Infrastructure Awareness

Introduction

When an unexpected alert fires, the clock starts ticking. Traditionally, engineers would ask an AI assistant for help, only to spend precious minutes explaining their environment—what services exist, which data sources hold metrics, how components connect. Each conversation starts from scratch. But with Grafana Assistant, you can eliminate that overhead entirely. Instead of learning your infrastructure on demand, Assistant proactively studies your environment and builds a persistent knowledge base. By the time you ask your first question, it already knows what's running, how it's connected, and where to look. This guide walks you through setting up that automated context preloading so you can jump straight into troubleshooting.

What You Need

• A Grafana Cloud stack (Grafana Assistant is included)
• At least one connected data source: Prometheus, Loki, or Tempo
• Permissions to enable AI features and view data source settings
• Basic familiarity with Grafana navigation

Step 1: Enable Grafana Assistant

First, ensure Grafana Assistant is active in your Grafana Cloud instance. Navigate to Administration > Plugins and search for “Grafana Assistant.” If it's not already enabled, click Enable. This activates the swarm of AI agents that will automatically analyze your data sources. No additional installation or complex configuration is needed—the service runs entirely in the background.

Step 2: Connect Your Data Sources

Assistant learns from the data sources you already have. Go to Connections > Data Sources and confirm that your Prometheus, Loki, and Tempo instances are listed and configured correctly. If any are missing, add them now. The agent swarm will use these connections to discover services, metrics, logs, and traces. For best results, ensure each data source is labeled clearly so Assistant can correlate them later.

Step 3: Let the Agent Swarm Work Its Magic

Once data sources are connected, Assistant starts its automatic discovery process. This requires zero input from you. Behind the scenes, a fleet of AI agents performs four key tasks:

1. Data source discovery – The system identifies all connected Prometheus, Loki, and Tempo data sources in your Grafana Cloud stack.
2. Metrics scans – Agents query Prometheus data sources in parallel to find services, deployments, and infrastructure components.
3. Enrichments via logs and traces – Loki and Tempo data sources are correlated with their corresponding metrics, adding context about log formats, trace structures, and service dependencies.
4. Structured knowledge generation – For each discovered service group, agents produce documentation covering five areas: what the service does, its key metrics and labels, how it's deployed, what it depends on, and where its logs and traces live.

This process runs continuously, so any new services or changes in your environment are automatically incorporated into the knowledge base.

Step 4: Review the Automatically Built Knowledge Base

After the initial scan (typically a few minutes), you can inspect what Assistant has learned. Open Grafana Assistant from the left sidebar and explore the generated knowledge base. You’ll see an organized view of your services, dependencies, and data source mappings. If something looks off, you can manually edit the knowledge entries—but for most setups, the automatic discovery is highly accurate. Think of this as giving Assistant a map of your world before you start asking questions.

Step 5: Start Asking Context-Aware Questions

Now the real benefit kicks in. When you ask “Why is my checkout service slow?” Assistant already knows that your payment system talks to three downstream services, that its latency metrics live in a specific Prometheus data source, and that its logs are structured JSON in Loki. You don’t need to explain anything. Conversations become faster and more accurate. This is especially powerful for teams where not everyone has the full infrastructure picture—developers can investigate upstream dependencies without prior knowledge.

Tips for Maximum Effectiveness

• Keep data sources clean – Label your Prometheus, Loki, and Tempo instances clearly so Assistant can correlate them correctly.
• Let the system warm up – The first scan may take a few minutes. After that, updates happen in near real-time as new services are deployed.
• Review generated documentation periodically – Check the knowledge base after major infrastructure changes to ensure accuracy.
• Combine with runbooks – Grafana Assistant can pull context from your existing runbooks if they are linked to data sources.
• Train your team – Encourage engineers to ask questions directly instead of manually searching for metrics. The more they use it, the faster incident response becomes.