Magnitude runs tests using Playwright. While agent.act() and agent.check() cover most test automation needs with natural language, direct access to Playwright’s page and context objects is available within test functions (async (agent) => { ... }).

Use agent.page and agent.context strategically when these specific capabilities are needed, or as a fallback if the AI struggles with a complex interaction. Refer to the Playwright Access reference for basic API details.

Magnitude aims to expand the capabilities of the ai agent over time to cover more scenarios natively. However, direct access to Playwright’s API ensures you always have the tools needed for any testing requirement in the meantime.

To configure the Playwright browser context, see browser options instead.

​
Example

import { test } from 'magnitude-test';

const mockProfileData = { userId: 'user123', notificationsEnabled: true };

test('can see correct user settings', async (agent) => {
  // Inject some authorization cookies directly with browser context
  await agent.context.addCookies([{ name: 'session_id', value: 'fake-session-token', domain: 'localhost', path: '/' }]);
  
  // Mock the settings API
  await agent.page.route('**/api/user/settings', async route => {
    await route.fulfill({ status: 200, contentType: 'application/json', body: JSON.stringify(mockProfileData) });
  });

  await agent.act("Navigate to the user settings page");
  await agent.check("Shows user ID 'user123'");
  await agent.check("Notifications are enabled");
});

By combining Magnitude’s AI-driven steps with direct Playwright access for these specific scenarios, you can create comprehensive and robust tests.