VERIFIED

RAG Web Browser

Safety: 65/100

0.0(0)

This MCP server enables AI agents to browse the web via the RAG Web Browser Actor, fetching and extracting content from web pages based on search queries or URLs.

MCP ServersSearch & WebAI & LLMDevelopmentData & DatabasesSocial MediaUtilitiesFreemium

Boost this tool

Subscribe to listing upgrades or segmented pushes.

Overview

This MCP server enables AI agents to browse the web via the RAG Web Browser Actor, fetching and extracting content from web pages based on search queries or URLs.

This MCP server is relatively safe for read-only web browsing. However, the ability to fetch arbitrary URLs and execute JavaScript introduces some risk. Ensure proper input validation and content sanitization to mitigate potential threats.

Performance depends on the speed of the RAG Web Browser Actor and the target websites. Consider using the raw-http scraping tool for faster performance on static pages.

Costs are associated with Apify's platform usage, including actor execution time and data storage. Be mindful of the number of requests and the size of the extracted content.

HybridJavaScript199 starsApache-2.0Released 8/21/2025CI/CD

Connections & Capabilities

Connects To

GitHubPlaywright

Capabilities

Read

Client Compatibility

Claude Desktop

full

Cursor

untested

Windsurf

untested

VS Code (Copilot)

untested

Quickstart

Install

npm install
npm run build

Claude Desktop Config

{
  "mcpServers": {
    "rag-web-browser": {
      "command": "npx",
      "args": ["@apify/mcp-server-rag-web-browser"],
      "env": {
        "APIFY_TOKEN": "your-apify-api-token"
      }
    }
  }
}

Environment Variables

APIFY_TOKEN

Tools (1)

moderate

search

Queries Google or fetches a URL, returning cleaned page content.

Can fetch arbitrary URLs and execute JavaScript, potentially exposing the agent to malicious content.

Security

Auth Method

API Key

Data Locality

hybrid

Known Risks

!Exposure to malicious websites through arbitrary URL fetching.
!Potential for denial-of-service attacks by requesting large or computationally intensive pages.
!Information leakage through extracted content.
!Dependency on the availability and security of the Apify platform.

Safety Score

moderate risk

Positive

+Limited to web browsing and content extraction.
+No direct system-level access.
+Uses Apify's sandboxed environment.
+Configuration options for request timeouts.

Risks

-Can fetch content from any URL, potentially exposing the agent to malicious sites.
-Depends on the security of the RAG Web Browser Actor.
-Potential for information leakage if extracted content contains sensitive data.
-The `scrapingTool` option allows JavaScript execution, increasing the attack surface.

Use Cases

Information Retrieval

-Answering questions based on web content.
-Summarizing web pages.
-Extracting data from websites.

Content Generation

-Generating blog posts based on web research.
-Creating product descriptions from competitor websites.
-Writing social media updates based on trending topics.

Research and Analysis

-Gathering data for market research.
-Analyzing competitor strategies.
-Tracking news and events.

Who Is This For?

Best For

AI agents that need to access and process web content.
RAG pipelines that require web browsing capabilities.
Tasks that involve information retrieval and summarization.
Applications where up-to-date information is crucial.

Not Ideal For

Tasks that require real-time interaction with websites.
Applications that handle sensitive data without proper security measures.
Scenarios where high reliability and availability are critical.

Autonomy & Execution

Default ModeRead-only

Sandboxed

The agent can autonomously browse the web and extract information, but it cannot perform any destructive actions.

StatelessYes

Retry LogicNot documented

ConcurrencySingle-threaded

Production Tip

Monitor Apify task usage and set appropriate request timeouts to prevent resource exhaustion.

FAQ

How do I authenticate with the MCP server?

You need to provide an Apify API token via the `APIFY_TOKEN` environment variable.

What scraping tools are available?

You can choose between `browser-playwright` (for JS-heavy sites) and `raw-http` (for faster scraping of static pages).

How can I limit the number of results returned by the search tool?

Use the `maxResults` argument to specify the maximum number of organic results to fetch.

What output formats are supported?

The server supports `text`, `markdown`, and `html` output formats.

How do I handle errors?

Errors are propagated back to the client via standard error. Implement error handling in your client application.

Can I use this server to scrape websites that require login?

Yes, but you need to configure the RAG Web Browser Actor to handle authentication.

Is this MCP server still supported?

This MCP server is deprecated. It is recommended to use mcp.apify.com or integrate directly with the RAG Web Browser Actor.

Key Features

Search & Web AI & LLM Development Data & Databases Social Media Utilities

Metrics

Discovered2/14/2026

Reviews0

Saved By0 users

Related Tools

RAG Web Browser | AI Builder Hub Review & Alternatives | Flaex AI