These notebooks provide a guided introduction to the core evaluation and guardrail features available in the Arthur Evals Engine (fka Arthur Shield). Use them to learn the workflow, test rules, and run performance checks with real examples.
A beginner friendly walkthrough that shows you how to create and delete default rules, tasks, and task based rules. This is the best place to start if you want a quick overview of how the Arthur Evals Engine handles evaluations and decision logic.
A collection of rule setups and test cases for every rule type supported in the Arthur Evals Engine. Each example shows how to define the rule, run checks, and interpret the results so you can adapt them to your own project.
A utility notebook for performance and scale testing. Use it to run large batches of prompts and responses through the Arthur Evals Engine, measure rule behavior at scale, and validate expected outcomes before production.