Aurora SRE
by Arvo AI
Open source AI agent for incident investigation and root cause analysis
Reviewed by the the30top research team · Last reviewed 2026-10-27
Overview
Aurora runs a LangGraph agent with more than 30 tools that pulls evidence from AWS, Azure, GCP, Kubernetes and monitoring platforms during an incident. It builds a knowledge graph of your infrastructure, searches runbooks and generates postmortems with Confluence export. It is self-hosted by default, so your team owns deployment, upgrades and model keys, which is real work for small teams.
Our verdict
A capable open source option for teams comfortable running their own stack. Expect setup effort and no vendor SLA.
Best for
Platform teams that want an open source incident agent they control
Avoid if
You want a fully managed product with a support contract
Pros
- Open source and self-hostable
- Wide tool coverage across clouds and monitoring
- Foreground chats ask for approval and background runs fail closed
Cons
- Self-hosting needs engineering time
- Young project with a small public track record
- LLM costs are yours to manage
Key capabilities
Frequently asked questions
Is Aurora SRE free?
Yes. It is Apache 2.0 open source and free to self-host. You cover your own model and infrastructure costs.
Does Aurora make changes without approval?
Foreground chats prompt for approval and background runs fail closed, though the vendor says it can run with no human in the loop if you choose.
Is customer data used for training?
The site states data is never used for model training.