KeelTest is a VS Code extension that generates comprehensive pytest suites by verifying code execution in real-time. Unlike standard LLM assistants, it does not guess; it creates a plan, generates valid python code, executes it in a local sandbox, and iteratively fixes failures before presenting the final output.
Most AI tools treat test generation as a text completion task. KeelTest treats it as an engineering problem.
Semantic Planning: Analyzes your project structure and dependencies to build a test specification before writing code.
The Triage Loop: Executes generated tests in your local environment immediately.
Auto-Correction: If a test fails, the agent analyzes the traceback, distinguishes between hallucinations and source bugs, and attempts to fix the test code automatically.
