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. This delivers 99% cost savings* and 10× faster runs compared to LLM-controlled agentic browsers.

*based on 1000 automations with different input criteria

Product tour
— What it is

Three steps from idea to scale.

Describe what you want in plain English. A3 helps you define the workflow, learns each site into Playwright stages, then runs them deterministically — with zero tokens at runtime.

"The Thinker" — judgment vs code.
01 · SETUP

Define the workflow.

Tell A3 what you want to automate. It asks clarifying questions, then proposes schemas, missions, and the steps your workflow should take on the site.

prompt · schemas · missions
02 · LEARN

Learn the site.

Point A3 at a target site. It navigates the funnel in the browser and builds verified Playwright stages one by one.

browser · stages · playbooks
03 · RUN

Run deterministically at scale.

Call learned workflows as an API on Automation Cloud, or expose them as MCP / UCP tools — no LLM in the loop.

API · MCP · UCP
— Works for any use case

Same 3 steps, different automation goals.

Hotel rates, flight booking, listings, checkout bots — the flow is identical: set up the workflow with A3, learn each site, then run at scale via API, MCP, or UCP. See the product tour for interactive demos.

"Vitruvian Man" — one model, any use case.
— Step 1 · Setup

Define the workflow.

  • Describe the automation in plain English.
  • A3 asks clarifying questions, then proposes schemas and the site steps you need.
— Step 2 · Learn

Learn the site.

  • Tell A3 which site to automate.
  • It navigates in the browser and writes verified Playwright stages one by one.
Waiting for stage…

        

        

          

            

              
              about:blank
            

          

          

            
Waiting for browser…

          

        

      

    

  






  

    

      

        — Step 3 · Run
        
Run deterministically at scale.

      

      
    
        
  • Create A3 jobs via the Automation Cloud API — 0 tokens, no LLM required.
    • 
        
  • Or expose the same project as MCP / UCP tools for agents.
    • 
      

    


    

      

        

          

            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
  }
}'

        
        
        

          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 →
        

      

    

  






  

    

      

        — Get started
        
Start building your first workflow in minutes.

        

          Begin with the quickstart in Docs and use the Product tour while implementing workflows and stages.
        

      

      

        Product tour