Skip to content

Concepts

Understanding the philosophy behind Tenro and when to use it.

This section explains the why behind Tenro's design. If you're looking for how to do something specific, see the Guides or API Reference.

Topics

  • How Tenro works

    How Tenro intercepts provider calls, why @link_llm is optional, and compatibility.

    Learn how it works

  • What is simulation?

    Understand the difference between simulation and traditional patching, and why it matters for AI agent testing.

    Learn about simulation

  • Testing strategy

    When to use Tenro vs. other testing approaches, and how to structure your test suite.

    Choose your strategy

  • Glossary

    Key terms and definitions used throughout the Tenro documentation.

    View glossary

Key principles

1. Simulation over patching

Traditional patching focuses on replacing functions. Tenro simulates tools, agents, and LLM behaviours (including errors) so your tests match production reality.

2. Test behaviour, not implementation

Verify that your agent does the right thing (calls the right tools, produces correct output) rather than how it's wired internally.

3. Single source of truth

Examples in examples/ are tested in CI and embedded in docs via snippets. Code you see here runs.