The power user's guide to the AI assistant in Grafana Cloud
Grafana Assistant can help you write queries, explain dashboards, summarize logs, and navigate your observability data. But that is only the starting point.
In this hands-on workshop, you’ll learn how to move from basic chat-based assistance to using Grafana Assistant as a powerful AI teammate for troubleshooting, automation, and operational knowledge sharing. You’ll start with everyday workflows like asking questions, finding dashboards, and understanding telemetry. Then you’ll progress into advanced patterns: using specialist capabilities, bringing business context into Assistant, creating reusable skills, running investigations, connecting external tools through MCP servers, and using Assistant from Slack, CLI, and other workflows.
By the end of the workshop, you’ll know how to make Assistant more useful, more accurate, and more aligned with the way your organization operates. You’ll leave with a practical blueprint for turning Assistant into a daily operational companion that helps your team investigate faster, standardize best practices, reduce repetitive toil, and scale observability expertise across your organization.
Whether you’re an SRE, platform engineer, developer, or observability leader, this workshop will help you turn Grafana Assistant from a useful helper into a force multiplier for your entire team.
What you'll accomplish
Use Grafana Assistant for everyday observability workflows such as queries, dashboards, logs, and troubleshooting
Apply specialist capabilities for dashboards and investigations
Bring your organization’s context into Assistant using rules, custom behaviors, memories, and reusable skills
Create skills that capture team expertise and standardize troubleshooting workflows
Connect Assistant to external tools with MCP servers, including issue trackers, code repositories, and project management systems
Use Assistant from Slack and CLI-based workflows to bring observability help where your team already works
Reduce repetitive operational toil with scheduled or workflow-based automation
Build a practical operating model for using Assistant across users, teams, and production incidents