Testkube
Kubernetes-native test orchestration with AI authoring and failure analysis.
About
Testkube turns your Kubernetes cluster into the test execution layer: you define test workflows in YAML, store them in Git, and Testkube runs them as native Kubernetes jobs inside your clusters, decoupled from whatever CI/CD tool you use. It is tool-agnostic, so Playwright, Cypress, k6, JMeter, Postman, pytest, OWASP ZAP, and anything container-based plug in, with results aggregated into one dashboard. The open-source agent is MIT-licensed and listed in the CNCF and CD Foundation landscapes, while the commercial control plane adds multi-cluster management and Testkube AI: plain-language workflow generation, automatic failure investigation, and remediation PRs.
The GitHub repository has 1,669 stars with commits landing the day of this writing. AI authoring is in early access and its quality claims are provider-reported, and running Testkube assumes a Kubernetes-native team. For platform teams drowning in plugin sprawl, a single testing layer that AI agents can query through MCP is a strong pitch.
What it can do
- Kubernetes-native test execution as native jobs
- YAML test workflows stored in Git
- Tool-agnostic: any container-based framework
- Testkube AI: workflow generation, failure investigation, remediation PRs (commercial)
- Multi-cluster control plane (commercial)
- MCP server for AI agent integration
- Native CI/CD plugins (Jenkins, GitHub Actions, GitLab, Argo)
- Centralized dashboard and artifact collection
Setup
- Install the Testkube agent into your cluster via Helm or CLI.
- Define test workflows in YAML and store them in Git.
- Connect your CI/CD tool via triggers or webhooks.
- Run workflows and review aggregated results; optionally trial the commercial control plane for Testkube AI.
Prerequisites
Kubernetes cluster (single cluster works for the open-source agent); container-based test tools.
Use cases
Works with
Trust score — 8/10
Scored 8/10 (TIER verified, metrics-v1). Evidence: 1,669 GitHub stars on kubeshop/testkube with a commit the day of writing (verified via GitHub API); open-source agent MIT-licensed; listed in CNCF and CD Foundation landscapes. Strengths: genuinely tool-agnostic, Kubernetes-native execution, healthy open-core model. Docked for: (1) AI authoring and remediation-PR features are early access with provider-reported quality; (2) commercial control plane is seat-based with quote-driven enterprise pricing. License: open core (agent MIT; control plane proprietary).
★ 1,669 GitHub stars
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 7.0/10
- Maintenance · recency, releases, issue load
- 9.0/10
- Openness · license + docs
- 8.0/10
- Manual review · sanity read + red-flag check
- 8.0/10
- · 1,669 GitHub stars on kubeshop/testkube with a commit the day of writing (GitHub API, 2026-10-16)
- · Open-source agent MIT-licensed; listed in CNCF and CD Foundation landscapes
- · Tool-agnostic Kubernetes-native execution; MCP server for AI agent integration
- · Testkube AI (workflow generation, failure investigation, remediation PRs) is early access with provider-reported quality
- · Commercial control plane is seat-based with quote-driven enterprise pricing
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.