Krkn-AI
github.com
- Free plan
- Linux
- Self-hosted
What Krkn-AI is
Krkn-AI is an open-source tool for running chaos scenarios and finding experiments that evaluate the resilience of Kubernetes and OpenShift applications. It uses evolutionary algorithms to generate experiments from Krkn scenarios, guided by user-defined goals such as SLOs and application health checks. Its discover command creates a YAML configuration populated with cluster component information and basic setup. Results show scenario effects on application liveness and include fitness scores; Prometheus metrics can help measure system response. An interactive dashboard monitors experiment results. The project requires a cluster kubeconfig and lists Python 3.11 or later, krknctl, oc or kubectl, Podman, and Helm 3.x among its prerequisites. It can run as a Python CLI or from a pre-built container image, including on Kubernetes. The repository lists an Apache-2.0 license. Its README describes the tool as under active development and advises users to use it at their own risk.
Who it suits
Krkn-AI is for teams evaluating the resilience of Kubernetes or OpenShift applications through chaos scenarios. It assumes access to a cluster and the listed command-line and deployment prerequisites.
Good to know, and what to watch
- Generates experiments with evolutionary algorithms.
- User-defined objectives can guide experiment generation.
- Reports application liveness impact and fitness scores.
- Prometheus metrics can measure system response.
- Available as a Python CLI or container image.
- Requires a Kubernetes or OpenShift cluster kubeconfig.
- Requires Python 3.11 or later and several listed tools.
- The README says the tool is under active development.
- The project advises users to use it at their own risk.
Plans
Every plan and priceApache License 2.0
Open source
Getting started
- Open the Krkn-AI GitHub repository
- Prepare Python 3.11 or later, krknctl, oc or kubectl, Podman, Helm 3.x, and a cluster kubeconfig
- Install the Python CLI from source or directly from the repository using uv
- Alternatively obtain the pre-built container image from Quay.io
- Run discovery to generate a YAML configuration with cluster details
- Execute experiments and review liveness, fitness scores, Prometheus measurements, or dashboard results
Limits to check
The facts list no quotas or feature limits for the free Apache-2.0 plan. Container use may require host networking and privileged mode in the standard example, although a FUSE-based option is documented.
The verdict
Krkn-AI combines scenario generation, user-defined evaluation goals, and reporting for Kubernetes and OpenShift resilience work. Review its prerequisites and active development status before incorporating it into a workflow.
Questions people ask
Is Krkn-AI free?
Yes. It is free, open-source software under the Apache-2.0 license.
Which platforms does it support?
It runs on Linux as a self-hosted tool and is intended for Kubernetes and OpenShift applications.
How can it be deployed?
You can install it as a Python CLI with uv or run a pre-built Quay.io container, including on Kubernetes.
What does it require?
Prerequisites include Python 3.11 or later, krknctl, oc or kubectl, Podman, Helm 3.x, and a Kubernetes or OpenShift cluster kubeconfig.
What does it measure?
It reports application liveness impact and fitness scores, and Prometheus metrics can measure system response.
What the maker publishes
- Algorithm
- It uses evolutionary algorithms to generate experiments based on Krkn scenarios. source
- Algorithm
- It uses evolutionary algorithms to generate experiments from Krkn scenarios and guide testing with user-defined objectives such as SLOs and application health checks. source
- Audience
- The project is intended for testing the resilience of Kubernetes and OpenShift applications. source
- Cluster discovery
- The discover command generates a YAML configuration pre-populated with cluster component information and basic setup. source
- Cluster discovery
- The discover command can populate configuration with cluster components, runnable scenarios, health check URLs, and Prometheus-validated fitness queries. source
- Container
- A pre-built container image is available on Quay.io, and the container guide documents discovery and run modes. source
- Container permissions
- The documented standard run example uses host networking and privileged mode; the guide also describes a FUSE-based option without the privileged flag. source
- Deployment
- Krkn-AI can be installed as a Python CLI or run from a pre-built container image, including on Kubernetes. source
- Development status
- The README says the tool is under active development and advises users to use it at their own risk. source
- Development status
- The repository cautions that the tool is under active development and should be used at the user's own risk. source
- Evaluation
- User-defined objectives such as SLOs and application health checks guide the algorithm. source
- Installation
- The README provides installation from source or directly from the GitHub repository using uv and Python 3.11 or later. source
- Integrations
- The README lists Prometheus metrics for measuring system response and requires krknctl and oc or kubectl. source
- Integrations
- The documented setup uses Krkn scenarios, krknctl, Kubernetes or OpenShift, and Prometheus metrics. source
- License
- The repository lists the Apache-2.0 license. source
- License
- The public repository is licensed under Apache-2.0. source
- Metrics
- Fitness evaluation can measure system response using Prometheus metrics. source
- Monitoring
- An interactive dashboard can monitor Krkn-AI experiment results. source
- Project affiliation
- The Krkn documentation identifies Krkn as a Cloud Native Computing Foundation Sandbox project. source
- Purpose
- Krkn-AI automatically runs chaos scenarios and discovers experiments to evaluate system resilience. source
- Purpose
- Krkn-AI automatically runs chaos scenarios and discovers experiments to evaluate system resilience. source
- Requirements
- The project lists Python 3.11+, krknctl, oc or kubectl, Podman, Helm 3.x, and a Kubernetes or OpenShift cluster kubeconfig as prerequisites. source
- Results
- The tool reports the impact of chaos scenarios on application liveness and their fitness scores. source
- Results
- Users can analyze scenario impact on application liveness and fitness scores, with an interactive dashboard for monitoring experiment results. source
Checked against the maker's pages on 10 Oct 2026.