# Grafana Labs Embeds Agentic AI Deeper into Analytics Platforms > Source: > Published: 2026-07-31 14:48:40+00:00 TL;DR — Key Takeaways - Grafana Labs introduced six AI capabilities spanning data discovery, agent monitoring, incident investigation, prompt automation and infrastructure management. - Its new MCP server and agentic tools aim to let users analyze observability data without needing to understand complex query languages. - AI-generated findings still require validation, particularly when they are used to automate critical business or operational workflows. Grafana Labs this week [added six artificial intelligence (AI) capabilities to its portfolio](https://www.businesswire.com/news/home/20260727216919/en/Grafana-Labs-Ships-Six-Tools-That-Power-Agentic-Operations-From-Planning-to-Production), including a Model Context Protocol (MCP) server that promises to make it simpler to discover, observe, and analyze a wide variety of different types of data. Launched during an AI Week event hosted by the company, other tools and capabilities include a Grafana Agent Observability tool that makes use of OpenTelemetry data to monitor AI agents, a Grafana Assistant Investigations tool, a tool that proactively forms the hypothesis for investigating incidents, Grafana Assistant Automations, a tool for reusing and scheduling prompts, and Grafana Assistant Workspace, a configuration tool that aggregates chat history, live canvas, and investigation report. Finally, there is also now an agentic command line interface (CLI) dubbed gcx, for managing dashboards, alert rules, data sources, and other resources as code. Collectively, these tools and capabilities are part of a larger effort to embed agentic AI capabilities more deeply across the Grafana Labs portfolio of data visualization and analytics tools and platforms, says Mat Ryer, senior director of AI at Grafana Labs. The overall goal is to make those tools and platforms more accessible to end users who often lack programming skills, he adds. “You no longer need to know the query language,” says Ryer. At the core of that capability are a set of harnesses and an orchestration engine that Grafana has added to its portfolio to make it simpler to create agentic AI workflows in a way that minimizes the total number of tokens that might otherwise need to be consumed, notes Ryer. As AI continues to evolve, it’s becoming more apparent that the level of expertise required to automate tasks is rapidly declining. As a result, it’s becoming simpler to more deeply analyze data to either surface a macro business intelligence trend or determine the root cause of an anomaly that might exist in a piece of code, says Ryer. While the degree to which an end user will be able to dive deeper into data will continue to vary depending on their level of expertise, the bar for applying advanced analytics is only going to continue to become lower, he adds. Of course, each end user will need to validate the output generated by any AI model, but there is a clear opportunity for organizations to make more fact-based decisions. Many of the processes that organizations have in place today are based on assumptions and biases that might not stand up to AI scrutiny. Additionally, more organizations should be able to use the validated output generated by AI tools to automate workflows. Depending on the level of criticality of those workflows, many of them might even be automated in a way that doesn’t always require human review. Regardless of approach, the one certain thing is that as AI continues to evolve, there will be no excuse for not considering many more potential facets before deciding that, if required, might not be so easily reversed as business workflows become more complex.