Technology · Ebook

HolmesGPT: Agentic Troubleshooting for Cloud-Native Systems

by Shriira Press

HolmesGPT is an open-source AI agent, built by Robusta with Microsoft and accepted into the CNCF Sandbox, that investigates production incidents and finds their root causes. Monitoring tells you that something is wrong; HolmesGPT works out why. Through an agentic loop, a language model decides which observability tools to call, pulls live data from Kubernetes, Prometheus, logs, Grafana and dozens of other sources, correlates the evidence, consults your runbooks, and writes a root-cause analysis. This book follows that idea from the problem of manual on-call debugging through the agentic loop, toolsets, runbooks, alert integrations with AlertManager, PagerDuty and Slack, the language models it runs on, and the CLI, server, and operator deployment modes — ending with the practices that make it trustworthy in production.

Contents

1. 1Preface
2. 2Chapter 1 — The Problem of the 3 a.m. Page
3. 3Chapter 2 — The Agentic Loop
4. 4Chapter 3 — Toolsets: Giving the Agent Senses
5. 5Chapter 4 — Runbooks and Knowledge
6. 6Chapter 5 — Alerts, Integrations, and Closing the Loop
7. 7Chapter 6 — Models and Deployment
8. 8Chapter 7 — HolmesGPT in Practice