Introducing simulation-based agent testing

Test your agents with scenarios directly in your codebase using Python and TypeScript, while being able to easily visualize results on LangWatch.

Test edge cases by simulating users in different scenarios. Then, let the simulation play out in multi-conversations — with no dataset required. 

From here, you can validate your full agent behavior: check for tool calls, handovers, integrate with guardrails, and more. 

Write as you would any other test, run on your CI/CD pipeline.

You can introduce a scalable and repeatable system that catches issues prior to production.

Get started: LangWatch Scenarios documentation →

Questions? Join our Discord community →