kubeshop/testkube
☸️ The Open Testing Platform for AI-Driven Engineering Teams
What it solves
Testkube is an open testing platform designed for engineering teams to define, run, and analyze automated tests directly within their Kubernetes infrastructure. It eliminates the need to build custom testing pipelines from scratch by providing a centralized way to execute any testing tool or script (API, E2E, Performance, Security, and Infrastructure tests) at scale.
How it works
The platform operates through an agent-based architecture. The open-source agent runs standalone in a Kubernetes cluster, allowing users to trigger tests manually, via schedules, CI/CD pipelines, GitOps, Kubernetes Events, or a REST API. It aggregates all test results, logs, and resource metrics for centralized troubleshooting. For more advanced orchestration across multiple clusters, a commercial control plane can be connected to these agents.
Who it’s for
It is built for AI-driven engineering teams and DevOps professionals who manage applications on Kubernetes and need a scalable, tool-agnostic testing framework that integrates with their existing scripts and tools.
Highlights
- Tool Agnostic: Supports any testing tool or script, including API, E2E, and performance tests.
- Flexible Triggering: Tests can be triggered via REST API, MCP, schedules, or Kubernetes Events.
- AI Integration: Includes an MCP server and native AI agents for troubleshooting, analysis, and remediation.
- Centralized Reporting: Aggregates logs, artifacts, and resource-metrics for easier troubleshooting.
- Deployment Options: Offers both a standalone MIT-licensed open-source agent and a commercial control plane for multi-cluster orchestration.