Futures
Access hundreds of perpetual contracts
TradFi
Gold
One platform for global traditional assets
Options
Hot
Trade European-style vanilla options
Unified Account
Maximize your capital efficiency
Demo Trading
Futures Kickoff
Get prepared for your futures trading
Futures Events
Join events to earn rewards
Demo Trading
Use virtual funds to experience risk-free trading
Launch
CandyDrop
Collect candies to earn airdrops
Launchpool
Quick staking, earn potential new tokens
HODLer Airdrop
Hold GT and get massive airdrops for free
Launchpad
Be early to the next big token project
Alpha Points
Trade on-chain assets and earn airdrops
Futures Points
Earn futures points and claim airdrop rewards
This is a beginner's tutorial for getting started with BrowserMCP.
🚀 Let AI Grow "Eyes" and "Hands": Beginner's Guide to BrowserMCP
Have you ever thought about how convenient it would be if your frequently used AI assistant (like Cursor or Claude Desktop) could not only write code and chat but also directly control your browser to look up information, fill out forms, or scrape web data?
BrowserMCP is the tool designed for this purpose. It acts as a bridge, connecting your AI assistant with the Chrome browser, allowing AI to "see" the world and perform operations through the browser.
🛠️ Preparation
Before starting, ensure you have installed the following software:
Node.js: The foundation for running the service (must be installed).
AI tools supporting MCP: Currently supporting Cursor, Claude Desktop, Windsurf, or VS Code.
Chrome browser (or Chromium-based browser).
⚡️ Three Steps to Quickly Get Started
Step 1: Configure the "Brain" (MCP Server)
First, we need to tell your AI tool how to connect to the BrowserMCP service. Here, using Cursor as an example (Claude Desktop is similar):
Open the Cursor settings panel.
Find the "Tools" (Tools) tab.
Click "New MCP server" (Create New MCP Server).
Enter the following information in the configuration box:
{
"mcpServers": {
"browsermcp": {
"command": "npx",
"args": ["@browsermcp/mcp@latest"]
}
}
}
Click save/refresh. The AI is now ready to send commands.
Step 2: Install the "Hands" (Browser Extension)
Next, we need to install a "receiver" in the browser to execute commands sent by AI.
Go to the BrowserMCP official website or extension store, and install the Browser MCP extension.
📌 Key Step: After installation, it’s recommended to pin (Pin) the extension icon to the browser toolbar for convenience.
Click the Browser MCP icon in the top right corner of your browser.
Click the “Connect” (Connect) button.
Note: This step connects the current tab to AI. All subsequent operations will be performed in this tab.
Step 3: Witness the Magic (Start Automation)
Everything is ready! Now, return to your AI tool (like Cursor's chat interface) and give commands as usual:
User: Open Google, search for "Browser MCP," and tell me what the first search result is.
What happens behind the scenes?
Cursor transmits your command to the local Server via MCP protocol.
The Server translates the instruction and sends it to the browser extension.
The browser extension automatically navigates to Google, inputs the text, and clicks search.
The extension reads the webpage content and sends it back to Cursor.
Cursor summarizes the content and replies to you.
💡 Tips for Use
What can it do?
“Help me summarize this webpage.”
“Check the latest issues on this GitHub project.”
“Help me fill out this test form (provide data).”
Now, your AI is no longer confined to the chat window. Go ahead and try to let it surf the internet! 🌊
Video demonstration