TestLLM

What it does. TestLLM lets you create scripted conversation sequences so your AI agents receive deterministic, repeatable responses during testing.

Why it matters. Real LLMs are non-deterministic, so the same prompt can produce different outputs and make E2E tests unreliable. TestLLM fixes this by acting as a drop-in LLM replacement that returns exact, scripted responses.

How it works. Define a test conversation with inputs and expected outputs, point your agent at TestLLM instead of a real LLM, and get the exact same behavior every time.