Browser Agents

AI-powered agents that autonomously complete browser tasks

What is a Browser Agent?#

A Browser Agent combines:

• Large Language Models (LLMs) for reasoning and decision-making
• Browser Sessions for executing actions
• Vision capabilities to understand web pages
• Autonomous planning to complete multi-step tasks

Unlike scripted automation, agents can adapt to changes, handle unexpected scenarios, and complete tasks without predefined workflows.

Quick Start#

from notte_sdk import NotteClient
 
client = NotteClient()
 
with client.Session(open_viewer=True) as session:
    agent = client.Agent(session=session, max_steps=5)
 
    response = agent.run(
        task="Browse on Notte docs and book a demo for me",
        url="https://docs.notte.cc"
    )
    print(response)

import { NotteClient } from 'notte-sdk';
 
const client = new NotteClient({
  apiKey: process.env.NOTTE_API_KEY,
});
 
await client.Session({ open_viewer: true }).use(async (session) => {
  const agent = client.Agent({ session, max_steps: 5 });
 
  const response = await agent.run({
    task: 'Browse on Notte docs and book a demo for me',
    url: 'https://docs.notte.cc',
  });
 
  console.log(response);
});

Tip: Agents run within browser sessions. Use context managers to ensure sessions are automatically stopped when done. This prevents orphaned sessions and unexpected costs.

How Agents Work#

1. Observation#

The agent observes the current page state: visible elements and their properties, interactive components (buttons, forms, links), text content and structure, current URL and page metadata.

2. Reasoning#

Using the LLM, the agent understands the current page, plans the next action, decides which element to interact with, and determines when the task is complete.

3. Action#

The agent executes browser actions: navigate to URLs, click buttons and links, fill forms, extract data, scroll and interact with dynamic content.

4. Iteration#

This cycle repeats until the task is successfully completed, the maximum steps are reached, or an error occurs that can't be resolved.

Agents vs Scripted Automation#

Both agents and scripted automation run on browser sessions, the cloud browser infrastructure. The difference is how you control what happens in that session.

Use scripted automation when: you know the exact steps to take, speed and cost are critical, or the target pages rarely change.

Use agents when: you don't know the exact steps, pages change frequently, or you need intelligent decision-making.

Note: You can combine both approaches: use an agent to figure out a workflow, then convert it to a function for faster, cheaper repeated execution.

Agent Capabilities#

• Structured Output — Get type-safe responses using Pydantic models

• Vaults & Personas — Use credentials and identities in automations

• Replay & Debugging — Debug with MP4 replays of agent execution

• Agent Fallback — Automatic recovery from script failures

Key Concepts#

Natural Language Tasks#

with client.Session() as session:
    agent = client.Agent(session=session)
    agent.run(task="Find the cheapest laptop under $1000 and add it to cart")

Structured Output#

from pydantic import BaseModel
 
class ContactInfo(BaseModel):
    email: str
    phone: str | None
 
with client.Session() as session:
    agent = client.Agent(session=session)
    result = agent.run(task="Extract contact information", response_format=ContactInfo)

Step Limits#

with client.Session() as session:
    agent = client.Agent(session=session)
    agent.run(
        task="Find and summarize the top 5 AI news from today",
        max_steps=20,  # Limit to 20 actions
    )

Error Handling#

agent = client.Agent(session=session)
result = agent.run(task="Complete task")
 
if result.success:
    print(result.answer)
else:
    print(f"Agent failed: {result.answer}")

Next Steps#

• Agent Lifecycle — Create, manage, and stop agents

• Agent Configuration — All configuration options

• Structured Output — Get typed responses from agents

• Convert to Functions — Turn agent runs into reusable code

Back to overview