Web automation that learns the site,
then runs it deterministically.

A3 uses LLMs to learn each site and writes Playwright scripts once — then the runner replays them with no model in the loop. 99% cheaper* and 10× faster than LLM-controlled agentic browsers.

*based on 1000 automations with different input criteria

Product tour

extract-hotels.stage.ts

idle

about:blank

loading page ...

— Setup

Define the workflow.

You describe the automation in plain English — what to do, on which sites, and what output you need.
A3 asks clarifying questions, then proposes schemas, missions, and the site steps your workflow should take.

— Learn

Learn the site.

You point A3 at a target site to learn.
A3 navigates the funnel in the browser and builds verified Playwright stages one by one into a playbook.

"Vitruvian Man" — one model, any use case.

Waiting for stage…

about:blank

Waiting for browser…

— Run

Run deterministically at scale.

You call learned workflows as jobs on the Automation Cloud API — 0 tokens, no LLM in the loop.
Or expose the same project as MCP / UCP tools so your agents can run them deterministically.

"Terracotta Warriors" — runners at scale.

Automation Cloud API Create a job

# Replay scrape-rates — no LLM in the loop
curl -X POST https://api.automation.cloud/v1/jobs \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
  "project": "scrape-rates",
  "site": "booking.com",
  "input": {
    "destination": "London",
    "checkIn": "2026-09-14",
    "checkOut": "2026-09-17",
    "guests": 2
  }
}'

const token = process.env.A3_TOKEN!;

const res = await fetch(
  'https://api.automation.cloud/v1/jobs',
  {
    method: 'POST',
    headers: {
      Authorization: `Bearer ${token}`,
      'Content-Type': 'application/json',
    },
    body: JSON.stringify({
      project: 'scrape-rates',
      site: 'booking.com',
      input: {
        destination: 'London',
        checkIn: '2026-09-14',
        checkOut: '2026-09-17',
        guests: 2,
      },
    }),
  },
);

const job = await res.json();
// deterministic playback · 0 tokens

import requests

requests.post(
  "https://api.automation.cloud/v1/jobs",
  headers={"Authorization": f"Bearer {token}"},
  json={
    "project": "scrape-rates",
    "site": "booking.com",
    "input": {
      "destination": "London",
      "checkIn": "2026-09-14",
      "checkOut": "2026-09-17",
      "guests": 2,
    },
  },
)

API Docs →

MCP · UCP Same project, agent-ready

MCP tool scrape_rates

Expose the learned workflow as a typed tool your agent can call.

"name": "scrape_rates",
"inputSchema": { … Rate search }

UCP endpoint POST /booking-com/ucp/mcp

Serve the same stages behind a UCP-compliant commerce contract.

tools/call → scrape_rates
# deterministic playback · 0 tokens

MCP & UCP docs →