mcp-server-rag-web-browser
- Browser Automation
Implementation of an MCP server for the RAG Web Browser Actor. This Actor serves as a web browser for large language models (LLMs) and RAG pipelines, similar to a web search in ChatGPT.
Prompts
Interactive templates invoked by user choice
Name | Description |
---|---|
search | Search phrase or a URL at Google and return crawled web pages as text or Markdown |
Resources
Contextual data attached and managed by the client
Name | Description |
---|---|
No resources |
Tools
Functions exposed to the LLM to take actions
Name | Description |
---|---|
search | Search phrase or a URL at Google and return crawled web pages as text or Markdown |
Server Configuration
Describes the environment variables required to run the server.
Name | Required | Description | Default |
---|---|---|---|
APIFY_API_TOKEN | Yes | Your Apify API token |
Model Context Protocol (MCP) Server for the RAG Web Browser Actor π
Implementation of an MCP server for the RAG Web Browser Actor. This Actor serves as a web browser for large language models (LLMs) and RAG pipelines, similar to a web search in ChatGPT.
π What is model context protocol?
The Model Context Protocol (MCP) enables AI applications (and AI agents), such as Claude Desktop, to connect to external tools and data sources. MCP is an open protocol that enables secure, controlled interactions between AI applications, AI Agents, and local or remote resources.
π― What does this MCP server do?
The RAG Web Browser Actor allows an AI assistant to:
- Perform web search, scrape the top N URLs from the results, and return their cleaned content as Markdown
- Fetch a single URL and return its content as Markdown
𧱠Components
Tools
- search: Query Google Search, scrape the top N URLs from the results, and returns their cleaned content as Markdown.
- Arguments:
query
(string, required): Search term or URLmax_results
(number, optional): Maximum number of search results to scrape (default: 1)
- Arguments:
Prompts
- search: Search phrase or a URL at Google and return crawled web pages as text or Markdown
- Arguments:
query
(string, required): Search term or URLmax_results
(number, optional): Maximum number of search results to scrape (default: 1)
- Arguments:
Resources
The server does not provide any resources and prompts.
π οΈ Configuration
Prerequisites
- MacOS or Windows
- The latest version of Claude Desktop must be installed (or another MCP client)
- Node.js (v18 or higher)
- Apify API Token (
APIFY_API_TOKEN
)
Install
Claude Desktop
Configure Claude Desktop to recognize the MCP server.
- Open your Claude Desktop configuration and edit the following file:
- On macOS:
~/Library/Application\ Support/Claude/claude_desktop_config.json
- On Windows:
%APPDATA%/Claude/claude_desktop_config.json
Copy"mcpServers": { "mcp-server-rag-web-browser": { "command": "npx", "args": [ "/path/to/mcp-server-rag-web-browser/build/index.js", ] "env": { "APIFY-API-TOKEN": "your-apify-api-token" } } } - On macOS:
- Restart Claude Desktop
- Fully quit Claude Desktop (ensure itβs not just minimized or closed).
- Restart Claude Desktop.
- Look for the π icon to confirm that the Exa server is connected.
- Examples You can ask Claude to perform web searches, such as:CopyWhat is an MCP server and how can it be used? What is an LLM, and what are the recent news updates? Find and analyze recent research papers about LLMs.
π·πΌ Development
Local Development
If you're working on an unpublished server, you can access the local server via the following command:
Local client
To test the server locally, you can use example_client
:
The script will start the MCP server, fetch available tools, and then call the search
tool with a query.
Debugging
Call the RAG Web Browser Actor to test it:
Since MCP servers operate over standard input/output (stdio), debugging can be challenging. For the best debugging experience, use the MCP Inspector.
Build the mcp-server-rag-web-browser package:
You can launch the MCP Inspector via npm
with this command:
Upon launching, the Inspector will display a URL that you can access in your browser to begin debugging.
GitHub Badge
Glama performs regular codebase and documentation scans to:
- Confirm that the MCP server is working as expected.
- Confirm that there are no obvious security issues with dependencies of the server.
- Extract server characteristics such as tools, resources, prompts, and required parameters.
Our directory badge helps users to quickly asses that the MCP server is safe, server capabilities, and instructions for installing the server.
Copy the following code to your README.md file: