Does Fetch execute JavaScript on pages?

No, it fetches standard HTTP responses and parses static HTML into markdown for high speed and low resource usage.

Web Fetch MCP Server Config Template

Target Standard & File Format

Select runtime specification

AI Agent Rules & Skills

Context Shields & Privacy (Ignore Files)

Integrations & Machine Documentation

AI Governance & Multi-Agent Planning Suite

MCP Server

Valid

Server key

Command

Arguments one per line

Env variable optional

Value

Valid for Claude Code (.mcp.json) and Claude Desktop (claude_desktop_config.json).

claude_desktop_config.json11 lines

{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "mcpServers": {
    "fetch": {
      "command": "uvx",
      "args": [
        "mcp-server-fetch"
      ]
    }
  }
}

Click or focus to edit

JSON•UTF-8•11 Lines•Spaces: 2

Target File Location

Project Placement

Save as .mcp.json in project root for Claude Code, or merge its mcpServers block into Claude Desktop claude_desktop_config.json.

What's in the ZIP export (20 files)

Unzip into your project root. Every file is already in the folder its tool reads from.

Include extras: Claude Code subagent, Gemini CLI, Aider, Copilot and Kiro scoped rules, CI audit workflow

.claude/settings.json
.claude/skills/codebase-auditor/SKILL.md
.claude/skills/codebase-auditor/scripts/verify.sh
.claudeignore
.cursor/rules/codebase-auditor.mdc
.cursorignore
.github/copilot-instructions.md
.mcp.json (open in editor)
.windsurf/rules/codebase-auditor.md
AGENTS.md
ARCHITECTURE.md
CLAUDE.md
DESIGN.md
MEMORY.md
PRD.md
README.md
TASK.md
llms.txt
prompts/gemini-system-instructions.json
prompts/openai-custom-instructions.md

What this file changes for Web Fetch & Markdown MCP

Without it

Fetches remote web documentation, API references, and HTML pages, converting them into clean token-efficient markdown for LLM comprehension.

What the file tells the agent

Uses `mcp-server-fetch` via uvx.

Converts raw HTML into clean, stripped markdown.

Enables the AI to read docs URLs provided in prompts.

How to Install Web Fetch & Markdown MCP Model Context Protocol (MCP) via Terminal

Step 1: Open Project Directory & Verify Target Placement

Open your terminal and navigate to your project root folder where the claude_desktop_config.json file will reside. Ensure the file is placed at the exact path below relative to your project root so the AI engine automatically loads it:

claude_desktop_config.json

Step 2: Fetch Rule File via Terminal Command

Run curl, PowerShell, or wget to stream the rule directly from the DevScratchpad raw API endpoint and write it to claude_desktop_config.json:

Terminal One-Liner Install

Run directly in your project root to stream and write this rule file with one command.

Raw API Stream

$curl -fsSL "https://www.devscratchpad.tech/api/raw/mcp-config/fetch" -o "claude_desktop_config.json"

Step 3: Check that Claude Code, Claude Desktop, Cursor uses it

Restart the client so it starts the server. Claude Code asks you to approve servers from .mcp.json; run /mcp to check their status.

Target Standard & File Format

MCP Server

Target File Location

Web Fetch & Markdown MCP Server Config

What this file changes for Web Fetch & Markdown MCP

How to Install Web Fetch & Markdown MCP Model Context Protocol (MCP) via Terminal

Step 1: Open Project Directory & Verify Target Placement

Step 2: Fetch Rule File via Terminal Command

Terminal One-Liner Install

Step 3: Check that Claude Code, Claude Desktop, Cursor uses it

MCP config generator: .mcp.json for Claude Code, Claude Desktop and Cursor

Tools for Web Fetch & Markdown MCP work

JSON Formatter & Validator

JWT Decoder & Inspector

JSON to TypeScript Interfaces

Diff Checker & Text Compare

Live Repository & Schema Ingestion

Migrating from Legacy .cursorrules or CLAUDE.md?

Technical FAQ: Web Fetch & Markdown MCP AI Rulebooks

Does Fetch execute JavaScript on pages?