MCP Server

Browser Use provides a hosted Model Context Protocol (MCP) server that enables AI assistants to control browser automation. Works with any HTTP-based MCP client, including Claude Code.

MCP Server URL: https://api.browser-use.com/mcp

This is an HTTP-based MCP server designed for cloud integrations and remote access. If you need a local stdio-based MCP server for Claude Desktop, use the free open-source version: uvx browser-use --mcp

Quick Setup

1. Get API Key

Get your API key from the Browser Use Dashboard

2. Connect Your AI

1
{
2
  "mcpServers": {
3
    "browser-use": {
4
      "command": "npx",
5
      "args": [
6
        "mcp-remote",
7
        "https://api.browser-use.com/mcp",
8
        "--header",
9
        "X-Browser-Use-API-Key: your-api-key"
10
      ]
11
    }
12
  }
13
}

Available Tools

The MCP server provides six tools:

browser_task

Creates and runs a full cloud browser automation task. Before using this tool, consider checking if a skill can accomplish the task faster (see list_skills).

• task (required): Detailed instructions for what you want the browser to do (max 50,000 characters)
• max_steps (optional): Maximum browser actions to take (1-100, default: 10)
• profile_id (optional): UUID of the cloud profile to use for persistent authentication
• Cost: 60 credits per step ($0.006/step) using Browser Use 2.0

list_skills

Lists all available pre-built automation skills. Skills execute in milliseconds without spinning up a browser, making them much faster than browser_task for supported actions.

• Parameters: None
• Returns: Array of skills with id, title, description, goal, and parameters
• Cost: Free

execute_skill

Executes a pre-built skill with provided parameters. For skills requiring authentication, first use get_cookies to extract the necessary cookie values.

• skill_id (required): UUID of the skill to execute (from list_skills)
• parameters (optional): Input parameters matching the skill’s schema. For type: "cookie" parameters, pass the cookie value from get_cookies
• Cost: 100 credits per execution ($0.01)

get_cookies

Extracts cookies from a browser profile for authenticated skill execution. Use this when a skill has parameters with type: "cookie".

• profile_id (required): UUID of the browser profile (from list_browser_profiles)
• domains (optional): Filter cookies by domain (e.g., ["x.com"])
• Returns: Array of cookies with name, value, domain, expires, and expires_in
• Cost: Free

list_browser_profiles

Lists all available cloud browser profiles for the authenticated project. Profiles store persistent authentication (cookies, sessions) for websites requiring login.

• Parameters: None
• Returns: Array of profiles with id, name, last_used_at, cookie_domains, and created_at
• Cost: Free

monitor_task

Gets current status and progress of a browser automation task, including detailed agent reasoning. Returns immediately with a snapshot - call repeatedly to track progress live.

• task_id (required): UUID of the task to monitor (returned by browser_task)
• Returns: Task status, step count, live_url, and for each step: goal, evaluation (agent reasoning), memory, actions, url, screenshot_url
• Cost: Free

Pricing

• Exchange rate: 10,000 credits = $1 USD
• Minimum balance: $0.10 (1,000 credits) required to execute skills

Example Usage

Once connected, ask your AI to perform web tasks:

“Search Google for the latest iPhone reviews and summarize the top 3 results”

“Go to Hacker News and get me the titles of the top 5 posts”

“Fill out the contact form on example.com with my information”

The AI will use the browser tools automatically to complete these tasks.

Using Skills (Fast Automation)

Skills are pre-built automation workflows that execute in milliseconds without spinning up a browser. Use skills when available for faster, more reliable results.

Recommended workflow:

1. Check for skills first: Call list_skills to see available automations
2. Check for cookie parameters: Look for parameters with type: "cookie" in the skill definition
3. Get cookies if needed: Call get_cookies with the profile_id and filter by the cookie’s domain
4. Execute the skill: Call execute_skill with the skill_id and parameters

Example: Posting to X/Twitter

1. list_skills → find "Post Tweet" skill with auth_token parameter (type: cookie, domain: x.com)
2. list_browser_profiles → get profile_id for your X account
3. get_cookies(profile_id="...", domains=["x.com"]) → extract auth cookies
4. execute_skill(skill_id="...", parameters={"tweet_text": "Hello!", "auth_token": "<cookie_value>"})

Smart Features

Cloud Profiles for Authentication

Use cloud browser profiles to maintain persistent login sessions across tasks. Profiles store cookies and authentication state for:

• Social media (X/Twitter, LinkedIn, Facebook)
• Email (Gmail, Outlook)
• Online banking and shopping sites
• Any website requiring login

List available profiles with list_browser_profiles, then pass the profile_id to browser_task.

Real-time Task Monitoring

Use monitor_task to check task progress while it’s running. The tool returns immediately with the current status, latest step details, and agent reasoning. Call it repeatedly to track progress live.

Conversational Progress Summaries

When you monitor tasks, the AI automatically interprets step data into natural language updates, explaining what the browser has completed and what it’s currently working on.

Troubleshooting

Connection issues?

• Verify your API key is correct
• Check you’re using the right headers

Task taking too long?

• Check the live_url to see real-time progress
• Increase max_steps for complex tasks (max: 100)
• Use clearer, more specific instructions
• Consider using skills instead of browser_task for supported actions

Skill execution failing?

• Ensure you have sufficient credits (minimum $0.10 balance)
• For authenticated skills, verify cookies are extracted with get_cookies first
• Check the skill’s required parameters match what you’re providing

Cookies not working?

• Verify the browser profile has an active session for the target site
• Check cookie expiration with the expires_in field
• Filter by the correct domain when calling get_cookies

Need help? Check the v2 API Reference for detailed specifications.