MCP NPX Fetch

MCP NPX Fetch is an MCP server that retrieves online resources and transforms web content into structured formats including HTML, JSON, Markdown, and plain text. Built with TypeScript, JSDOM, and Turndown, it connects AI clients like Claude Desktop directly to external web addresses. Developers, researchers, and automation engineers use it to pull live documentation, ingest public API endpoints, extract article text, and strip away unnecessary styling and scripts. By running over standard Model Context Protocol interfaces, the server exposes dedicated fetching tools that can pass custom HTTP headers for authenticated endpoints. This setup enables language models to read remote documentation in readable Markdown, parse raw responses, or inspect clean text without needing external browser automation platforms or manual copy-pasting.

Category: Browser & Web Automation

Tags: data-extraction, fetch, html, markdown, web scraping

Visit MCP NPX Fetch

How to install and configure MCP NPX Fetch

Follow these steps to install and register MCP NPX Fetch with Claude Desktop: 1. Open your Claude Desktop configuration file: - macOS: ~/Library/Application Support/Claude/claude_desktop_config.json - Windows: %APPDATA%/Claude/claude_desktop_config.json - Linux: ~/.config/Claude/claude_desktop_config.json 2. Add the server under the mcpServers object: json { "mcpServers": { "fetch": { "command": "npx", "args": ["-y", "@tokenizin/mcp-npx-fetch"], "env": {} } } } 3. Alternatively, install it globally using npm install -g @tokenizin/mcp-npx-fetch and run mcp-npx-fetch. 4. Restart Claude Desktop to apply the configuration and access the fetching tools.

What you can do with MCP NPX Fetch

• Converting web articles and online documentation into clean Markdown for summarization within LLM prompts. * Fetching and parsing remote JSON data from REST APIs using custom HTTP headers for authenticated endpoints. * Extracting stripped plain text from web pages while eliminating HTML tags, scripts, and embedded styling. * Retrieving raw HTML documents from remote URLs for detailed DOM inspection, debugging, and web parsing tasks.

Key facts

• https://github.com/tokenizin-agency/mcp-npx-fetch
• Browser & Web Automation, Files, Documents & PDFs, Web Search & Research
• data-extraction, fetch, html, markdown, web scraping

Part of MCP Servers

What is MCP NPX Fetch?

MCP NPX Fetch is a Model Context Protocol server developed by PT Tokenizin Technology Agency. It provides AI agents with tools to fetch web content and automatically transform it into formats such as plain text, Markdown, JSON, or raw HTML.

How do I install MCP NPX Fetch?

You can run it on demand using npx with the package name @tokenizin/mcp-npx-fetch, or install it globally across your system using npm install -g @tokenizin/mcp-npx-fetch. It can also be added directly to your Claude Desktop configuration.

What tools are available in MCP NPX Fetch?

The server provides four core tools: fetch_html for retrieving raw HTML, fetch_json for fetching and parsing JSON, fetch_txt for clean plain text stripped of tags, and fetch_markdown for converting pages to Markdown format. All tools support custom HTTP headers.

Which MCP clients work with MCP NPX Fetch?

It is fully compatible with Claude Desktop as well as any other client application that supports the Model Context Protocol specification over stdio.

Is MCP NPX Fetch open source?

Yes, MCP NPX Fetch is an open source project distributed under the MIT License. You can view, clone, or contribute to the source code on GitHub.