Testkube launches AI Test Creation to turn plain-language prompts into runnable tests inside teams’ infrastructure

# What happened Testkube introduced a feature called AI Test Creation that converts plain-language test descriptions into working tests, integrates them into existing pipelines, and runs them inside a customer's infrastructure before the team accepts them.

# Why this matters Teams are shipping more code, and a growing portion is drafted with AI. Test creation and wiring into pipelines has lagged, so coverage falls behind commits. Testkube's approach focuses on the full lifecycle: generation, wiring, execution, and versioning. That reduces the gap between AI-written code and tested software by making tests immediately tangible and reviewable within teams' existing workflows.

# Creation does

• Generates tests in whatever framework a team already uses, across end-to-end, API, load, infrastructure testing, and more. Skills are built for widely used tools and scenarios.
• Executes every generated test inside the team's real environment within seconds so wrong assumptions surface while the test is still a draft.
• Creates pull requests in the team's GitHub repository so tests can be reviewed, edited, and versioned like any other code.
• Supports on-premises deployment so tests and execution data can remain inside the customer's cluster and use LLMs the customer provides.

# How Testkube positions the feature Testkube frames the problem as one of follow-through: creating a test is only half the work. The rest—wiring it into repositories and pipelines, and proving it in a real environment—is where teams often fail to keep pace. Ole Lensmar, Testkube co-founder and CTO, said AI widening the gap between committed code and testing was the motivation for building AI Test Creation. The product is designed to integrate with existing tools and provide immediately integrated tests that are proven in real infrastructure.

# Practical implications for teams

# What the feature does not claim The announcement focuses on integration and workflow automation. It does not claim universal correctness of generated tests, nor does it promise elimination of human review—tests still arrive as pull requests for teams to accept, edit, and version.

# Where to start

# Bottom line AI Test Creation addresses the operational gap that forms after a test is created: implementation, environment wiring, execution, and repository integration. The feature emphasizes immediate execution in real environments and repository-based review to keep tests aligned with committed code.