Sign in Book a demo
Back to Tools
Developer Tools API Key BYOC
Build Your Agent View Bright Data Docs
brightdata logo

Bright Data MCP Tools for AI Agents

Let an agent search the web, scrape pages as Markdown, and extract structured data from sites like LinkedIn, Amazon, and Instagram.

Available Tools

6 tools available in this integration

Brightdata.ExtractWebPageFields

Extract structured data from various websites like LinkedIn, Amazon, Instagram, etc.

Brightdata.GetWebPage

Scrape a webpage and return content in Markdown format using Bright Data.

Brightdata.ScrapeAsMarkdown

Scrape a webpage and return content in Markdown format using Bright Data.

Brightdata.SearchEngine

Search using Google, Bing, or Yandex with advanced parameters using Bright Data.

Brightdata.SearchWebPages

Search using Google, Bing, or Yandex with advanced parameters using Bright Data.

Brightdata.WebDataFeed

Extract structured data from various websites like LinkedIn, Amazon, Instagram, etc.

Why agent-optimized tools matter

Built for intent, not API schemas

Ask the agent to "compare the latest pricing pages for five competitors and return the source URLs." Arcade picks the right Bright Data tools without guessing which pages to search or scrape first.

Up to 100x fewer tokens

Bright Data searches return focused result records, while page content is fetched only when needed, so the agent can choose what to read instead of loading whole sites into the model.

Complete the workflow end to end

Search the web, open relevant pages, extract fields, compare findings, and return structured results through one toolkit designed around the complete web research task.

Make actions predictable and controlled

Arcade tools return structured results and useful errors across users and runs. Teams can apply scope restrictions, redaction, and approval policies so agents follow the same controls every time.

Securing Bright Data

The more capabilities agents get, the more useful they are. However, more capabilities bring more risks of undesired effects such as data loss and exfiltration.

This is why Arcade provides you with contextual rules that dictate how and when a tool is allowed to run. Arcade’s contextual access lets you govern tool visibility and behavior at every stage of execution and control who can see a tool, what inputs are allowed, and what comes back.

Learn about Contextual Access Policies

For Bright Data, you can implement some of these policies:

Access when tools are listed for a user
  • Decide which tools an agent is allowed to see at all. A tool that is never listed cannot be called, whoever is asking.
Pre-execution before each tool call
  • Inspect the inputs a call is about to run with, and block it or send it for human approval when they fall outside what the agent should do unsupervised.
Post-execution after the tool returns
  • Redact API keys, session tokens, email addresses, and phone numbers from Brightdata.ScrapeAsMarkdown results before the model sees scraped page content.
  • Strip email addresses, phone numbers, and exposed credentials from Brightdata.SearchEngine snippets and result text before returning them to the model.
  • Deny Brightdata.WebDataFeed results containing private profile details or personal contact fields, so the model cannot receive individuals' non-public data.

How to use Bright Data with your agent

Follow the walkthrough to connect Bright Data to your MCP client through Arcade. Then give your agent tools for web search, Markdown scraping, and structured web data.

Your questions, answered

A Bright Data MCP server exposes web search, page retrieval, scraping, and structured data extraction as tools an AI agent can call through the Model Context Protocol. It lets an agent find pages, read their content, and extract fields from sites such as LinkedIn, Amazon, and Instagram.

Bright Data's own MCP server provides direct access to its web data capabilities. Arcade adds the runtime around those calls: a shared credential is brokered securely, and teams can apply approval, redaction, and access policies from any supported client.

Arcade agents can search Google, Bing, or Yandex; retrieve web pages as Markdown; scrape pages as Markdown; and extract structured data from supported sites such as LinkedIn, Amazon, and Instagram. These Bright Data actions are read-only, so they do not publish, edit, or delete content on the sites they access.

Arcade uses a Bright Data API key or secret configured for the integration. That credential needs access to the Bright Data products and data services your enabled workflows use; this integration does not use per-user OAuth scopes.

No. Bright Data uses one shared API key or secret configured once for the integration, rather than having each user connect a separate account. Arcade brokers the credential, and the model does not receive it.

Yes. Contextual Access policies can require human approval before any Bright Data call, including a broad search or scrape that you want a person to review first. Because all actions here are read-only, there are no Bright Data actions that change or delete external data.

Arcade proxies Bright Data API calls and does not persist your content as part of normal tool execution. The API key or secret is brokered and stored encrypted at rest, and the model never receives the raw credential.

Any framework or client that can connect to Arcade's MCP gateway or tool APIs works, including LangChain, OpenAI Agents, CrewAI, Claude Desktop, Claude Code, Cursor, Visual Studio Code, Microsoft Copilot Studio, and GitHub Copilot.

Arcade cannot refresh a shared Bright Data API key after it is revoked, rotated, or otherwise becomes invalid. Calls stop until the current key or secret is configured again.

Bright Data's account- and product-specific rate limits, concurrency limits, and quotas still apply to calls made through Arcade. The applicable limits depend on the Bright Data services and plan behind the configured credential.

Ready to actually ship your agent?

Connect your agent to Bright Data with a shared API key, agent-optimized tools and approval controls for consequential actions.

Build Your Agent Read the Bright Data Quickstart