Project Arena

Project Arena is a vendor-neutral Kubernetes incident benchmark for testing AI investigation tools.

Project Arena provides a Kubernetes fixture to inject faults, save investigation records from any product, and score completed investigations using a configurable judge. Users can choose a local kind cluster or AWS EKS, and select either a 21-scenario full suite or a six-scenario smoke fixture. The tool supports direct deployment with kubectl or a GitOps path using Argo CD. It compares native AI investigations from products like Edge Delta and Grafana, as well as external models like Claude using observability CLIs. The benchmark evaluates detection, root cause analysis, blast radius, supported final mitigation, and implementation readiness.

Project Arena at a glance

Best for: Software engineer
Runs on: Linux, macOS, Windows, CLI, self-hosted, open source
Pricing: Free
Released: October 2026
Category: AI research tool
Website: github.com/edgedelta/project-arena

Key features

Deploy a disposable Kubernetes fixture
Inject faults into the cluster
Save investigation records from any product
Score completed investigations with a configurable judge
Support local kind or AWS EKS clusters
Run a 21-scenario full suite or six-scenario smoke fixture
Support direct deployment or GitOps with Argo CD

Use it for

Test AI investigation tools across detection, diagnosis, and mitigation
Compare AI incident investigation performance across different platforms
Evaluate root cause analysis and blast radius accuracy
Measure implementation readiness of proposed mitigations

Alternatives to Project Arena

Meow AI Arena — Access complete match materials, original AI code, and local challenge scripts.
Fightbench — It tests artificial intelligence models on game memory rather than pixels.
SlopGrid — Create your own AI model benchmark and compare public grids.
SAG — It combines model routing, knowledge management, and agent orchestration in private.
Kwatch — Traces every cluster failure to its root cause and sends one message.
AntV Chart MCP — Provides direct integration of diverse chart generation tools into AI applications.
Artificial Analysis — Independent benchmarks for AI models across quality, speed, and cost.
Navital Gym — Measure AI agent performance on realistic tasks using deterministic rules without LLM judges.
Lacuna by Tiptree Systems — Traces code imports and comments directly to the foundational research papers.
UltraRAG — Build complex RAG workflows using modular MCP components and YAML configuration.

All Project Arena alternatives