Bright Data MCP
Bright Data's Web MCP server enabling AI agents to search, extract & navigate the web. It exposes 32 tools covering web search, page scraping, structured data extraction from 1,200+ pre-built scrapers, browser automation, and LLM response collection. Every request is routed through Bright Data's unblocking infrastructure to handle bot detection, CAPTCHAs, and geo-restrictions.
Package and access
Local package with source-attributed installation metadata. Koskamo does not execute packages during review.
What you can do with it
- Search Google, Bing, and Yandex for structured results
- Scrape any URL as Markdown or HTML with automatic proxy rotation
- Extract clean JSON from 1,200+ pre-built scrapers like Amazon and LinkedIn
- Automate browser navigation, clicking, and screenshotting
Source and maintenance
- Maintenance checked
Packages and installation
Source-provided package data. Koskamo has not executed or live-probed this package.
@brightdata/mcp 4.1.1
{
"version": "4.1.1",
"transport": "stdio",
"identifier": "@brightdata/mcp",
"registryType": "npm",
"registryBaseUrl": "https://registry.npmjs.org",
"environmentVariables": [
{
"name": "API_TOKEN",
"format": "string",
"isRequired": true,
"description": "Your API key for Bright Data"
},
{
"name": "WEB_UNLOCKER_ZONE",
"format": "string",
"description": "Your unlocker zone name"
},
{
"name": "BROWSER_ZONE",
"format": "string",
"description": "Your browser zone name"
},
{
"name": "GROUPS",
"format": "string",
"description": "Comma-separated tool group IDs (e.g., 'ecommerce,browser')"
},
{
"name": "TOOLS",
"format": "string",
"description": "Comma-separated tool names to enable"
},
{
"name": "BASE_TIMEOUT",
"format": "number",
"description": "Request timeout in seconds for base tools (search & scrape). No limit if unset"
},
{
"name": "BASE_MAX_RETRIES",
"format": "number",
"description": "Max retries for base tools on transient errors (0-3, default: 0)"
}
]
}Tools
Reported by the attributed source; not independently observed by Koskamo.
Connect from your client
Each snippet configures the same source-provided package; replace placeholders before use.
Cursor
{
"mcpServers": {
"bright-data-mcp": {
"command": "npx",
"args": [
"-y",
"@brightdata/mcp@4.1.1"
],
"env": {
"API_TOKEN": "<YOUR_API_TOKEN>",
"WEB_UNLOCKER_ZONE": "<YOUR_WEB_UNLOCKER_ZONE>",
"BROWSER_ZONE": "<YOUR_BROWSER_ZONE>",
"GROUPS": "<YOUR_GROUPS>",
"TOOLS": "<YOUR_TOOLS>",
"BASE_TIMEOUT": "<YOUR_BASE_TIMEOUT>",
"BASE_MAX_RETRIES": "<YOUR_BASE_MAX_RETRIES>"
}
}
}
}Cline
{
"mcpServers": {
"bright-data-mcp": {
"command": "npx",
"args": [
"-y",
"@brightdata/mcp@4.1.1"
],
"env": {
"API_TOKEN": "<YOUR_API_TOKEN>",
"WEB_UNLOCKER_ZONE": "<YOUR_WEB_UNLOCKER_ZONE>",
"BROWSER_ZONE": "<YOUR_BROWSER_ZONE>",
"GROUPS": "<YOUR_GROUPS>",
"TOOLS": "<YOUR_TOOLS>",
"BASE_TIMEOUT": "<YOUR_BASE_TIMEOUT>",
"BASE_MAX_RETRIES": "<YOUR_BASE_MAX_RETRIES>"
}
}
}
}VS Code
{
"servers": {
"bright-data-mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@brightdata/mcp@4.1.1"
],
"env": {
"API_TOKEN": "<YOUR_API_TOKEN>",
"WEB_UNLOCKER_ZONE": "<YOUR_WEB_UNLOCKER_ZONE>",
"BROWSER_ZONE": "<YOUR_BROWSER_ZONE>",
"GROUPS": "<YOUR_GROUPS>",
"TOOLS": "<YOUR_TOOLS>",
"BASE_TIMEOUT": "<YOUR_BASE_TIMEOUT>",
"BASE_MAX_RETRIES": "<YOUR_BASE_MAX_RETRIES>"
}
}
}
}Claude Desktop
{
"mcpServers": {
"bright-data-mcp": {
"command": "npx",
"args": [
"-y",
"@brightdata/mcp@4.1.1"
],
"env": {
"API_TOKEN": "<YOUR_API_TOKEN>",
"WEB_UNLOCKER_ZONE": "<YOUR_WEB_UNLOCKER_ZONE>",
"BROWSER_ZONE": "<YOUR_BROWSER_ZONE>",
"GROUPS": "<YOUR_GROUPS>",
"TOOLS": "<YOUR_TOOLS>",
"BASE_TIMEOUT": "<YOUR_BASE_TIMEOUT>",
"BASE_MAX_RETRIES": "<YOUR_BASE_MAX_RETRIES>"
}
}
}
}Let your agent find its tools
Find servers like Bright Data MCP from your agent. Add Koskamo once to search the index for endpoints, access requirements, and client setup. Public discovery needs no key.