Week of June 20–26, 2026
The headline this week is Contexts & Scenarios — the most significant change to how LitmusLab works since we started.
Contexts & Scenarios 🔥
Most AI features don't operate in a vacuum. They operate on data: a user's account history, their recent activity, their preferences, their current state. That data shapes what the AI can say — and whether what it says is actually correct.
Contexts let you define the structure of that data. If your AI has access to a user's watch history and friend list, your Context captures that shape. Scenarios are real instances of it — the specific users, accounts, or states you want to test against.
Once you've set this up, every test case runs against every scenario automatically. Five test cases, four user scenarios: twenty results, each evaluated against the right data for that user. The Scenario Matrix shows you at a glance where your AI works across users and where it breaks down.
This changes what's possible in testing. Instead of a single made-up input, you test against data that reflects how your AI actually operates in production.
API Requests
We renamed "Steps" to "API Requests" throughout the app. Each step in a test is an HTTP request to your API — the new name says exactly what it is.
