krkn-chaos/krkn ? reverse-engineered prompt

Build me a Python tool for chaos and resiliency testing in Kubernetes.

I want it to let me run planned failure tests against a cluster, like killing pods, disrupting nodes, and other stressful situations, so I can see whether my apps recover well and keep performing. Make it feel like a real project I can install, configure, and run from the command line, with support for test scenarios and clear configuration files. If it makes sense, include a small local server or API so I can trigger runs and check results. Add the basic pieces needed for Kubernetes access, permissions, and sample scenarios, plus enough docs so someone can get started without reading the code first.

Keep it practical and focused on reliability testing, not just a toy demo. Use current Python best practices and look up any recent Kubernetes or packaging details online if you need to.