Web

Brightdata MCP adapted for Heroku deployment
  • javascript

1

GitHub Stars

javascript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

The Web MCP lets you give LLMs and AI agents real-time web access by enabling search, navigation, and data extraction from the live web, while handling bot defenses and geo restrictions. It provides a streamlined way to run web-enabled agents and retrieve up-to-date information without getting blocked.

How to use

You will connect your MCP client to either a remote MCP server URL or a local MCP runner, then issue web-related queries through your AI assistant. Start by configuring a remote MCP URL to allow your agent to access the live web or run a local MCP node via a standard command. The MCP handles web requests, unlocks site navigation, and returns structured data that your agent can consume. When you need to run tasks that require browsing, data extraction, or real-time updates, the MCP provides the relevant endpoints and tools to perform those actions securely.

How to install

Prerequisites: you need Node.js installed on your system to run MCP commands via npx, and you should have an API token from Bright Data to authorize web access.

Step-by-step setup using the HTTP remote MCP (recommended for quick start):

  1. Obtain your API token from Bright Data and keep it handy.

How to install

{
  "mcpServers": [
    {
      "type": "http",
      "name": "brightdata_mcp",
      "url": "https://mcp.brightdata.com/mcp?token=YOUR_API_TOKEN_HERE",
      "args": []
    },
    {
      "type": "stdio",
      "name": "brightdata_mcp_local",
      "command": "npx",
      "args": ["@brightdata/mcp"],
      "env": [{"name": "API_TOKEN", "value": "YOUR_API_TOKEN_HERE"}]
    }
  ]
}

Additional sections

Security best practices: Treat web-scraped data as untrusted. Filter and validate data before using it in prompts, and prefer structured data extraction over raw text to minimize prompt injection risks.

Account setup: You need an active Bright Data account with an API key. Use the token in your MCP config and environment where required.

Deployment notes: You can run the MCP server locally with npx @brightdata/mcp or connect to the remote MCP URL. For production, consider setting rate limits and dedicated Web Unlocker and Browser zones as described in the configuration steps.

Troubleshooting: If you encounter timeouts, increase the timeout in your agent configuration. If the system cannot find npx, specify the full path to Node in the command (for example, "/usr/local/bin/node" as the command) and retry.

Notes on usage and configuration

Environment and tokens: You will typically pass API_TOKEN as an environment variable or include it in the remote MCP URL. Ensure you keep tokens secure and avoid exposing them in logs.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational
Web MCP Server - dsouza-anush/brightdata-mcp-heroku | VeilStrat