Skip to main content
Anthropic’s Claude models with Computer Use capabilities enable AI agents to interact with web browsers through screenshots and native computer control actions.

​
Overview

Anthropic Computer Use (anthropic-cua) leverages Claude’s vision and reasoning capabilities to:
  • Analyze screenshots of web pages
  • Identify interactive elements
  • Execute precise mouse and keyboard actions
  • Complete complex multi-step workflows

​
Supported Models

ModelModel IDBest For
Claude Opus 4.5claude-opus-4-5-20251101Most advanced reasoning, complex workflows (default)
Claude Sonnet 4.5claude-sonnet-4-5-20250929Balanced performance, general automation
Claude Opus 4claude-opus-4-20250514Advanced coding and agentic tasks
Claude Sonnet 4claude-sonnet-4-20250514Fast, reliable automation

​
Code Example

import Anchorbrowser from 'anchorbrowser';

const anchorClient = new Anchorbrowser({
  apiKey: process.env.ANCHORBROWSER_API_KEY
});

const response = await anchorClient.agent.task(
  'Navigate to GitHub and find the anchorbrowser repository',
  {
    taskOptions: {
      url: 'https://github.com',
      agent: 'anthropic-cua',
      // model: 'claude-opus-4-5-20251101',  // Default model
      maxSteps: 30
    }
  }
);

console.log(response);

​
Configuration Options

ParameterTypeDescription
agentstringMust be anthropic-cua
modelstringClaude model to use (default: claude-opus-4-5-20251101)
urlstringStarting URL for the task
max_stepsintegerMaximum actions the agent can take
output_schemaobjectJSON Schema for structured output

​
Best Practices

  • Opus 4.5 is the default - most capable model for complex workflows
  • Use Sonnet 4.5 for faster automation when speed is important
  • Provide clear, specific prompts describing the exact task to complete