Web Content MCP

What is Web Content MCP?

Web Content MCP is an MCP server that provides two tools – web-search (Google search plus scraping top results) and url (direct URL content retrieval). It is designed for users of Claude Desktop, VSCode, or any MCP-compatible agent who need quick, automated access to web content.

How to use Web Content MCP?

Clone the repository, install dependencies with pnpm install, and create a .env file with GOOGLE_API_KEY and GOOGLE_SEARCH_ENGINE_ID. Configure your MCP client (Claude Desktop or VSCode) with a stdio server using the command pnpx tsx pointing to src/index.ts and the .env file. Invoke via prompts such as “Search for information about [query]” or “Visit and summarize information from [URL]”.

Key features of Web Content MCP

• Fast and efficient web content retrieval
• Search-based and direct URL access
• Compatible with Claude Desktop, VSCode, any MCP agent
• Uses Google Custom Search API for queries
• Automatically scrapes content from visited pages

Use cases of Web Content MCP

• Searching Google and summarizing the top results
• Directly fetching and scraping content from a specific URL
• Gathering research data without manual browsing
• Integrating live web data into AI agent workflows

FAQ from Web Content MCP

What tools does Web Content MCP provide?

It provides web-search (searches Google and scrapes top results) and url (directly visits provided URLs and scrapes content).

What API keys are required?

You need a GOOGLE_API_KEY and a GOOGLE_SEARCH_ENGINE_ID stored in a .env file in the server’s root directory.

How do I set up Web Content MCP in Claude Desktop?

Add a JSON entry under mcpServers with "command": "pnpx", "args" including "tsx", the --env-file flag pointing to your .env, and the full path to src/index.ts.

How do I set up Web Content MCP in VSCode?

In your VSCode settings.json, add an MCP server entry with "type": "stdio", "command": "pnpx", and the same "args" as Claude Desktop, ensuring paths are properly escaped on Windows.

What should I do if I encounter connection or API errors?

Verify that the server is running and paths in your configuration are correct. Check that your GOOGLE_API_KEY and GOOGLE_SEARCH_ENGINE_ID are valid in the .env file.