# web-fetch-mcp

This MCP server lets an agent read web pages, summarize them, compare URLs, and pull out requested details.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 9
- Author: KylinMountain
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/KylinMountain/web-fetch-mcp
- Updated: 2025-09-18
- HTML: https://affpapa.org/en/nevibecode/mcp/web-fetch-mcp

## What it does

web-fetch-mcp exposes three tools for web research: `summarize_web` creates a short account of linked pages, `compare_web` checks information across URLs, and `extract_web` finds requested details from page content. A single request can include up to 20 URLs. HTML is converted into readable text, and GitHub `/blob/` links are resolved to raw content.

The server runs through `npx` and connects over MCP. It requires `GEMINI_API_KEY`; `HTTP_PROXY` and `HTTPS_PROXY` can be supplied when the route to Gemini needs a proxy. Without `GEMINI_API_KEY`, the server will not start. Requests to private IP addresses are blocked, while timeout and content-size limits can be configured.

## Requirements
- Needs: Node.js (commands run via npx)
- Environment variables: `GEMINI_API_KEY`, `HTTP_PROXY`, `HTTPS_PROXY` — keys and tokens: ask the human for the values, never commit them or put them in shared files.

## Install

### Claude Code (native platform)

**Platform CLI**
```bash
claude mcp add --env GEMINI_API_KEY=<GEMINI_API_KEY> --env HTTP_PROXY=<HTTP_PROXY> --env HTTPS_PROXY=<HTTPS_PROXY> --transport stdio web-fetch-mcp -- npx -y web-fetch-mcp
```
Replace <…> with your own values

### Codex

**Platform CLI**
```bash
codex mcp add web-fetch-mcp --env GEMINI_API_KEY=<GEMINI_API_KEY> --env HTTP_PROXY=<HTTP_PROXY> --env HTTPS_PROXY=<HTTPS_PROXY> -- npx -y web-fetch-mcp
```
Replace <…> with your own values

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "web-fetch-mcp": {
            "command": "npx",
            "args": [
                "-y",
                "web-fetch-mcp"
            ],
            "env": {
                "GEMINI_API_KEY": "<GEMINI_API_KEY>",
                "HTTP_PROXY": "<HTTP_PROXY>",
                "HTTPS_PROXY": "<HTTPS_PROXY>"
            }
        }
    }
}
```
Replace <…> with your own values

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "web-fetch-mcp": {
            "type": "stdio",
            "command": "npx",
            "args": [
                "-y",
                "web-fetch-mcp"
            ],
            "env": {
                "GEMINI_API_KEY": "<GEMINI_API_KEY>",
                "HTTP_PROXY": "<HTTP_PROXY>",
                "HTTPS_PROXY": "<HTTPS_PROXY>"
            }
        }
    }
}
```
Replace <…> with your own values

### Gemini CLI

**Platform CLI**
```bash
gemini mcp add -e GEMINI_API_KEY=<GEMINI_API_KEY> -e HTTP_PROXY=<HTTP_PROXY> -e HTTPS_PROXY=<HTTPS_PROXY> web-fetch-mcp npx -y web-fetch-mcp
```
Replace <…> with your own values

## When to use it
- summarize research pages by URL
- compare claims across several sites
- extract named details from web content
- process up to 20 URLs in one request

## Example prompts
- Use `summarize_web` to give me the main points from these URLs: ...
- Call `compare_web` and contrast the pricing and terms on these pages: ...
- Use `extract_web` to find publication dates and author names in these pages: ...

## Similar
- [nansen-web-fetcher](https://affpapa.org/en/nevibecode/skills/nansen-web-fetcher.md) — skill · 457 installs
- [pulse-fetch](https://affpapa.org/en/nevibecode/mcp/pulse-fetch.md) — MCP server · ★ 81
- [Defuddle Fetch](https://affpapa.org/en/nevibecode/mcp/defuddle-fetch.md) — MCP server · ★ 8
- [baoyu-danger-gemini-web](https://affpapa.org/en/nevibecode/skills/baoyu-danger-gemini-web.md) — skill · 27900 installs
- [web-access-claude-skill](https://affpapa.org/en/nevibecode/skills/web-access-claude-skill.md) — skill · 965 installs
- [CRW Web Scraper](https://affpapa.org/en/nevibecode/mcp/crw-web-scraper.md) — MCP server · ★ 1181

---
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