⬢github Python · 7 ★ +5 since we first saw it · pushed 6 d ago · MIT
edgedelta/project-arena
A vendor-neutral Kubernetes incident benchmark for testing AI investigation tools across detection, diagnosis, and mitigation. Deploy faults, capture investigations, and compare results with a configurable AI judge.
Project Arena is an open benchmark for evaluating AI SRE/investigation tools on Kubernetes. It spins up a disposable cluster (kind or EKS), injects faults across 21 incident scenarios, lets you connect any observability product to investigate, records the investigations, and scores them with a configurable AI judge so vendors and agents can be compared head-to-head.
Why now: It was shared on Hacker News as a Show HN with published benchmark results comparing Edge Delta, Grafana, and Claude-driven investigations across the same 21 scenarios.
Who it is for: SRE/platform teams and observability vendors who want to objectively evaluate AI incident-investigation tools on Kubernetes.
Stars over our 29 snapshots: 2 to 7, since 7 h ago.
Where people talked about it
API: https://socialmediatrends-api.osmike.com/v1/repos/edgedelta/project-arena