Up and running in 5 minutes
From zero to your first AI-generated, eval-gated test in under 5 minutes. No config files. No boilerplate. Just describe what you want tested.
Create your account
Sign up for a free Spectr account. You get $1 of free Claude credit to generate your first tests — no credit card required.
Create accountInstall the Spectr SDK
Add the Spectr SDK to your project. It works alongside your existing Playwright or pytest setup.
npm install @spectr/sdk --save-dev
Configure your API key
Copy your API key from the Settings page and add it to your environment.
# .env.local SPECTR_API_KEY=sk-spectr-xxxxxxxxxxxx
Generate your first test
Describe what you want to test in plain English. Spectr generates a Playwright spec, scores it through the eval pipeline, and streams the result back.
import { Spectr } from '@spectr/sdk';
const s = new Spectr({ apiKey: process.env.SPECTR_API_KEY });
const result = await s.generate({
story: 'User logs in with valid credentials and lands on the dashboard',
baseUrl: 'https://app.myproduct.com',
});
console.log(result.evalReport.gate); // 'pass' | 'warn' | 'fail'
console.log(result.code); // Playwright specRun in CI
Add the eval gate to your GitHub Actions workflow to block releases when test quality drops below your threshold.
# .github/workflows/ci.yml
- name: Spectr eval gate
run: npx spectr eval --strict
env:
SPECTR_API_KEY: ${{ secrets.SPECTR_API_KEY }}View CI docsExplore the full SDK reference, CI integration guides, and eval pipeline docs.