Brightdata

The Web MCP lets you give AI assistants real-time internet access through a dedicated MCP server. It enables seamless online research, web browsing, and data extraction for AI workflows while handling access and unblocking concerns so your AI can fetch current information reliably.

How to use

You connect your MCP client to a remote MCP endpoint or run a local MCP server to enable real-time web access for your AI. Use the hosted URL when you want a zero-install setup and manage your API key in the client configuration. If you prefer running locally, you provide the MCP command and environment token in a configuration file, then start the MCP alongside your AI client.

How to install

Prerequisites: ensure you have Node.js installed on your machine. You can verify by running node -v and npm -v in your terminal.

Option 1 — Use the hosted MCP URL (no installation required): add this MCP URL to your MCP client configuration: https://mcp.brightdata.com/mcp?token=YOUR_API_TOKEN_HEREIn Claude Desktop or other compatible clients, add a new custom connector with the following details:

• Name: Bright Data Web
• URL: https://mcp.brightdata.com/mcp?token=YOUR_API_TOKEN
• Save to complete the setup.```

Option 2 — Run a local MCP server (requires Node/NPX): create a configuration file with the following content, then start the MCP client as instructed by your tooling:

{ "mcpServers": { "Bright Data": { "command": "npx", "args": ["@brightdata/mcp"], "env": { "API_TOKEN": "<your-api-token-here>" } } } }

Save this as a JSON configuration file (for example, `mcp_config.json`) and ensure your MCP client is pointed to it. The token placeholder should be replaced with your actual API token before use.

Additional setup notes

If you plan to enable advanced features, you can run with Pro mode by setting environment variables as described in the advanced configuration. This unlocks additional tools and capabilities such as browser automation and web data APIs when you need deeper web interactions.

Available tools

search_engine

AI-optimized web search returning Google-like results for research and timely information

scrape_as_markdown

Converts any webpage into clean Markdown suitable for documentation or content extraction

scraping_browser.*

Full-featured browser automation for advanced interactions (Pro mode)

web_data_*

Web data API tools for structured data extraction and integration