Grafana Assistant Embraces AI Agents for Proactive Observability and Automation
Grafana Labs introduces six new AI-powered capabilities for Grafana Assistant, transforming it into an agentic operations layer for automated detection, investigation, and remediation of production issues.
Grafana Labs has announced the general availability of six AI capabilities, significantly expanding Grafana Assistant into an agentic operations layer designed to proactively detect, investigate, and remediate production issues. These new features aim to shift the focus of observability earlier in the software development lifecycle, addressing the increasing pace of code deployment and the growing complexity of modern systems.
The enhancements include Grafana Assistant Investigations, Grafana Assistant Workspace, and Grafana Assistant Automations, alongside updates to the Grafana Cloud MCP server, gcx, and Grafana Agent Observability. This suite of tools is intended to help engineers manage the challenges of rapid development by providing insights and automation from the planning stages through to production.
According to Grafana Labs’ 2026 Observability Survey, a significant majority of practitioners see value in AI for anomaly detection, yet many are still in the early stages of implementing observability for their own AI systems. Grafana Assistant aims to bridge this gap by delivering high-quality insights from production telemetry precisely when they are needed, not just after a failure occurs.
"We used to treat observability as something you bolt on just before code reaches production," said Mat Ryer, Senior Director of AI, Grafana Labs. "That’s changing. Now, Grafana Assistant can review your plans before you’ve written a line of code, add the instrumentation for you once you have, watch features as they land in production, and stay with you as you scale while dealing with the inevitable incidents that follow. It’s the same assistant end-to-end, and it doesn’t just advise; it acts."
Grafana Assistant Workspace, now generally available, provides a centralized hub for pre-development planning, allowing engineers to bring architecture designs into the tool to identify potential scaling issues before any code is written. When paired with Grafana Assistant Investigations, ongoing investigations can be transformed into shareable reports without manual rewriting.
For the transition from planning to running code, Grafana Assistant assists in instrumenting new services by automatically creating pull requests with the necessary instrumentation, configuring data sources, and verifying telemetry. This process is powered by gcx, Grafana’s agentic CLI for managing resources as code, and the Grafana Cloud MCP server, which provides direct access to Grafana instances for AI agents.
In production, Grafana Assistant Investigations and Automations help engineers quickly understand telemetry data by allowing plain-language queries and correlating technical metrics with business outcomes. Investigations can form hypotheses about incidents and systematically chase down leads in the data, while Automations can be scheduled to run recurring checks or prompts, sending summaries to channels like Slack without manual intervention.
Finally, Grafana Agent Observability extends Grafana Cloud’s monitoring capabilities to AI systems themselves. This feature, born from the need to observe Grafana Assistant's own usage, captures standard telemetry like usage, latency, and errors, alongside crucial AI-specific metrics such as token usage, cost, and conversation data, ensuring the reliability and efficiency of the AI agents being deployed.