🔥 FireScrape MCP Server

The Ultimate Gateway for AI Agents. Empower your LLMs to Browse, Search, Research, and Scrape with precision.

Installation • Connection Guides • AI Rules • Tools

🚀 Overview

FireScrape MCP is a Model Context Protocol server designed to bridge the gap between AI models (Claude, Cursor, etc.) and the real world. It provides robust tools for:

• Deep Research: Multi-step autonomous investigation and synthesis.

• Web Access: Anti-blocking search and intelligent markdown scraping.

• Local Context: File system access for project-aware coding.

Related MCP server: WebRead MCP

📦 Installation & Usage

Run instantly without global installation using npx:

npx -y @yutugyutugyutug/firescrape-mcp

✨ Key Features (v1.1.2)

• 🧠 deep_research: Conducts autonomous, multi-step research on complex topics, filtering noise and synthesizing reports.

• 🌐 search_web: High-availability web search with fallback providers (DuckDuckGo + Brave) to ensure you always get results.

• 📄 scrape_webpage: Converts messy HTML into clean, token-efficient Markdown, stripping ads and irrelevant boilerplate.

• 🔍 discover_tools: a "Meta-Tool" to find and install other MCP servers from GitHub and Smithery.

• 📂 Local Integration: Seamlessly list_files and read_file to give your AI context about your local project.

🤖 AI Agent Rules

Want your AI to be smarter? We have created a standardized rule set to teach your AI agents (Cursor, Windsurf, Cline) how to effectively use these tools to verify facts and avoid hallucinations.

👉 Read the AI Usage Guidelines

Copy the content of the file above into your .cursorrules or System Prompt to supercharge your agent.

🔌 Connection Guides

1. Claude Desktop App

1. Open your configuration file:

   • Windows: Win + R -> %APPDATA%\Claude\claude_desktop_config.json

   • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

2. Add this configuration:

   {
     "mcpServers": {
       "firescrape": {
         "command": "npx",
         "args": ["-y", "@yutugyutugyutug/firescrape-mcp"]
       }
     }
   }

2. Cursor (AI Code Editor)

1. Navigate to Settings -> Features -> MCP.

2. Click + Add New MCP Server.

3. Fill in the fields:

   • Name: FireScrape

   • Type: stdio

   • Command: npx

   • Args: -y @yutugyutugyutug/firescrape-mcp

4. Click Add to save.

3. Gemini CLI

1. Locate your config file:

   • Linux/Mac: ~/.gemini/settings.json

   • Windows: %USERPROFILE%\.gemini\settings.json

2. Insert the server configuration:

   {
     "mcpServers": {
       "firescrape": {
         "command": "npx",
         "args": ["-y", "@yutugyutugyutug/firescrape-mcp"]
       }
     }
   }

4. Claude Code (CLI)

Run this command in your terminal to add FireScrape:

claude mcp add firescrape -- npx -y @yutugyutugyutug/firescrape-mcp

5. Google Antigravity

1. Open the Manage MCP Servers panel.

2. Select "Edit Configuration" or open mcp_config.json.

3. Add the entry:

   "firescrape": {
     "command": "npx",
     "args": ["-y", "@yutugyutugyutug/firescrape-mcp"]
   }

6. Cline (VS Code Extension)

1. Open Cline Settings (in VS Code settings or via the extension menu).

2. Locate the MCP Servers config file (usually via a button "Edit MCP Settings").

3. Add the following:

   {
     "mcpServers": {
       "firescrape": {
         "command": "npx",
         "args": ["-y", "@yutugyutugyutug/firescrape-mcp"]
       }
     }
   }

🛠️ Tools Reference