Skip to content

Automation Cheatsheet

Here’s a cheatsheet—a quick overview and reference—for the Webfuse Automation API.

Extension and Automation API Namspaces

Hierarchical diagram of Webfuse APIs and the Automation API

Perceive the Current Page

// DOM-based
await browser.webfuseSession
  .automation
  .see
  .domSnapshot()

// Vision-based
await browser.webfuseSession
  .automation
  .see
  .guiSnapshot()

Act in the Current Page

// DOM-based (by CSS selector)
await browser.webfuseSession
  .automation
  .act
  .click("button#submit")

// Vision-based (by point coordinate)
await browser.webfuseSession
  .automation
  .act
  .click([374, 1020])

Downsample a Perceived DOM Snapshot

// DOM-based
const snapshot = await browser.webfuseSession
  .automation
  .see
  .domSnapshot()

const snapshot32KTokens = await browser.webfuseSession
  .tool
  .downsample(snapshot);

Downsampling is a technique to ‘compress’ a DOM snapshot:

<section class="container" tabindex="3" required="true" type="example">
  <div class="mx-auto" data-topic="products" required="false">
    <h1>Our Pizza</h1>
    <div>
      <div class="shadow-lg">
        <h2>Margherita</h2>
        <p>
          A simple classic: mozzarela, tomatoes and basil.
          An everyday choice!
        </p>
        <button type="button">Add</button>
      </div>
      <div class="shadow-lg">
        <h2>Capricciosa</h2>
        <p>
          A rich taste: mozzarella, ham, mushrooms, artichokes, and olives.
          A true favourite!
          </p>
        <button type="button">Add</button>
      </div>
    </div>
  </div>
</section>

…could be downsampled to:

<section tabindex="3" class="container" required="true">
  # Our Pizza
  <div class="shadow-lg">
    ## Margherita
    A simple classic mozzarela tomatoes and basil
    <button type="button">Add</button>
    ## Capricciosa
    A rich taste A true favourite
    <button type="button">Add</button>
  </div>
</section>

The Agent and Browser Ecosystem

	Webfuse	Playwright	Puppeteer
Human-like (natural) pointer	✔️	❌	❌
Human-like (natural) typing	✔️	✔️	✔️
Cross-frame targeting	✔️	❌	❌
Cross-shadow targeting	✔️	Manual	Manual
GUI snapshots	✔️	✔️	✔️
DOM snapshots	✔️	✔️	✔️
DOM downsampling	✔️	❌	❌
Accessibility tree	✔️	Partial	❌
LLM-based agents focus	✔️	❌	❌
Remote browser	❌	✔️	✔️
Embedded/user browser	✔️	Stream	Stream
MCP-ready	✔️	Separate	Separate

Example Agent Extension Repositories