Browser Agent

🚀 New: Magnitude is state-of-the-art, scoring 94% on WebVoyager!

Magnitude uses vision AI to enable you to control your browser with natural language.

• 🧭 Navigate - Sees and understands any interface to plan out actions
• 🖱️ Interact - Executes precise actions using mouse and keyboard
• 🔍 Extract - Intelligently extracts useful structured data
• ✅ Verify - Built-in test runner with powerful visual assertions

You can use it to automate tasks on the web, integrate between apps without APIs, extract data, test your web apps, or as a building block for your own browser agents.

↕️ Magnitude in action! ↕️

// Magnitude can handle high-level tasks
await agent.act('Create a task', {
    // Optionally pass data that the agent will use where appropriate
    data: {
        title: 'Use Magnitude',
        description: 'Run "npx create-magnitude-app" and follow the instructions',
    },
});

// It can also handle low-level actions
await agent.act('Drag "Use Magnitude" to the top of the in progress column');

// Intelligently extract data based on the DOM content matching a provided zod schema
const tasks = await agent.extract(
    'List in progress tasks',
    z.array(z.object({
        title: z.string(),
        description: z.string(),
        // Agent can extract existing data or new insights
        difficulty: z.number().describe('Rate the difficulty between 1-5')
    })),
);

Get started

Running your first browser automation

npx create-magnitude-app

This will create a new project and walk you through the steps for setting up Magnitude. It will also create an example script that you can run right away!

Using the test runner

To install the test runner for use in an existing web app, please run:

npm i --save-dev magnitude-test && npx magnitude init

This will create a basic tests directory tests/magnitude with:

• magnitude.config.ts: Magnitude test configuration file
• example.mag.ts: An example test file

For information on how to run tests and integrate into CI/CD see here.

!NOTE] Magnitude requires a large **visually grounded** model. We recommend Claude Sonnet 4 for the best performance, but are also compatible with Qwen-2.5VL 72B. See [docs for more information.

Why Magnitude?

❌ Problem #1: Most browser agents draw numbered boxes around page elements - doesn't generalize well due to complex modern sites

✅ Solution: Vision-first architecture

• Visually grounded LLM specifies pixel coordinates
• True generalization independent of DOM structure
• Future-proof architecture for desktop apps, VMs, etc.

❌ Problem #2: Most browser agents follow "high-level prompt + tools = work until done" - works for demos, not production