# Defuddle Fetch

Downloads a web page by URL and gives the agent clean markdown with title, author and date. Handy when the agent has to read docs or an article.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 8
- Author: domdomegg
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/domdomegg/defuddle-fetch-mcp-server
- Updated: 2026-10-21
- HTML: https://affpapa.org/en/nevibecode/mcp/defuddle-fetch

## What it does

The server adds a single tool, `fetch`. Give it a `url` and it downloads the page, strips the surrounding clutter with Defuddle and hands back the main text as markdown, together with title, author, publication date, word count, domain and processing time.

It is built as a replacement for the stock fetch server that relies on Readability, and the author says Defuddle copes better with most modern pages. Set `raw` to true if you want the page content without markdown conversion.

Only 5000 characters come back by default, so a long article has to be read in slices: `start_index` sets the offset and `max_length` the slice size. It launches through `npx`, which means Node.js must be installed.

## Requirements
- Needs: Node.js (commands run via npx)

## Install

### Claude Code (native platform)

**Platform CLI**
```bash
claude mcp add defuddle-fetch-mcp-server -- npx -y defuddle-fetch-mcp-server
```

### Codex

**Platform CLI**
```bash
codex mcp add defuddle-fetch -- npx -y defuddle-fetch-mcp-server
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "defuddle-fetch": {
            "command": "npx",
            "args": [
                "-y",
                "defuddle-fetch-mcp-server"
            ]
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "defuddle-fetch": {
            "type": "stdio",
            "command": "npx",
            "args": [
                "-y",
                "defuddle-fetch-mcp-server"
            ]
        }
    }
}
```

### Gemini CLI

**Platform CLI**
```bash
gemini mcp add defuddle-fetch npx -y defuddle-fetch-mcp-server
```

## When to use it
- read a docs page or article from a link inside the agent chat
- pull the author and publication date out of a page
- read a long page in slices by moving `start_index`
- get the original HTML when markdown is the wrong format
- swap the stock Readability-based fetch server for a cleaner one

## Example prompts
- Use `defuddle-fetch` to read the article at the link above and tell me who wrote it and when
- Fetch the next slice of the same page with `fetch`, starting at index 5000
- Grab this page through `defuddle-fetch` with `raw` on, I need the HTML

## Similar
- [Scrapling MCP Server](https://affpapa.org/en/nevibecode/mcp/scrapling-mcp-server.md) — MCP server · ★ 89433
- [browser-use](https://affpapa.org/en/nevibecode/mcp/browser-use.md) — MCP server · ★ 119160
- [AIHawk](https://affpapa.org/en/nevibecode/mcp/aihawk.md) — MCP server · ★ 32221
- [Skyvern](https://affpapa.org/en/nevibecode/mcp/skyvern.md) — MCP server · ★ 23263
- [SeleniumBase MCP](https://affpapa.org/en/nevibecode/mcp/seleniumbase-mcp.md) — MCP server · ★ 13169
- [apify-mcp-server](https://affpapa.org/en/nevibecode/mcp/apify-mcp-server.md) — MCP server · ★ 12927

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md