Skip to main content
Want to know what OpenSRE actually looks like when you run it? Here’s a walkthrough of a first install, the files you get after an investigation, and a way to put your own demos on the page.

Watch a first run

Install, set things up, and throw a sample alert at it. Most people are done in under ten minutes.
Running an OpenSRE investigation on a sample alert through to a structured root-cause report
Prefer step-by-step with platform-specific GIFs? Start at the Quickstart.

Full quickstart

Install, onboard, and run the sample alert — with screenshots for macOS, Linux, and Windows.

Jump to the GIFs

Skip straight to the install and onboard recordings.

What you get after a run

When an investigation finishes, OpenSRE drops a few files you can open in any editor: Need one file for a script or pipeline instead?
You can also send the summary to Slack. Here’s what that looks like:
Slack incident summary after an OpenSRE investigation
More on the full flow, slash commands, and delivery options: Investigations overview.

Put it through CloudOpsBench

Curious how OpenSRE stacks up? CloudOpsBench is a 452-scenario Kubernetes RCA suite — you can run OpenSRE against it and compare to published LLM-only baselines.
Setup, integrity checks, and cost notes live on the CloudOpsBench page.

Show us what you built

Got a demo, a runbook, or a post-mortem that used OpenSRE?
  1. Post in GitHub Discussions with the Showcase tag — include the commands you ran, which tools were connected, and redacted output if you can.
  2. Join the bi-weekly giveaway when you share something — there’s OpenSRE swag in the mix.