nicolagaletti/welcome-to-jungle-mcp-server

Welcome to the Jungle MCP Server

A Model Context Protocol (MCP) server that enables Claude to search and access job listings from Welcome to the Jungle using Puppeteer-based web scraping.

Features

• 🔍 Search jobs by keywords, location, and contract type
• 📋 Get detailed information about specific job postings
• 🤖 Seamless integration with Claude Desktop
• 🆓 No API keys or external services required - uses direct web scraping
• 🚀 Built with Puppeteer for reliable headless browser scraping
• ✅ Comprehensive unit test coverage

Prerequisites

• Node.js 18 or higher
• Claude Desktop app

Installation

1. Clone this repository:

   git clone https://github.com/yourusername/welcome-to-jungle-mcp-server.git
   cd welcome-to-jungle-mcp-server
   

2. Install dependencies:

   npm install
   

3. Configure Claude Desktop:

   Edit your Claude Desktop config file:

   • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
   • Windows: %APPDATA%\Claude\claude_desktop_config.json

   Add this configuration:

   {
     "mcpServers": {
       "welcome-to-jungle": {
         "command": "node",
         "args": ["/absolute/path/to/welcome-to-jungle-mcp-server/index.js"]
       }
     }
   }
   

   Replace /absolute/path/to/welcome-to-jungle-mcp-server/index.js with the actual path to your installation.

4. Restart Claude Desktop

Usage

Once configured, you can ask Claude to help with job hunting:

Example Prompts

• "Search for software engineer jobs in Paris"
• "Find remote product manager positions"
• "Show me internships in London"
• "Get details about this job: [paste URL]"
• "Compare these three job postings and help me decide which to apply for"

Available Tools

search_jobs

Search for jobs on Welcome to the Jungle.

Parameters:

• query (required): Job title, company name, or keywords
• location (optional): City or "Remote"
• contractType (optional): e.g., "full_time", "internship", "freelance"
• maxResults (optional): Number of results (default: 20)

get_job_details

Get detailed information about a specific job posting.

Parameters:

• jobUrl (required): Full URL of the job posting

How It Works

This server uses Puppeteer, a headless Chrome browser, to scrape Welcome to the Jungle's job listings directly. Unlike API-based approaches:

• ✅ No API keys or external accounts needed
• ✅ No usage limits or quotas
• ✅ Works with client-side rendered React applications
• ✅ Handles dynamic Algolia-based search results
• ⚠️ First request may take a few seconds as Chrome launches

Development

Run tests:

npm test

Project structure:

• index.js - MCP server entry point
• src/scraper.js - Puppeteer web scraping logic
• src/handlers.js - Tool request handlers
• src/tools.js - Tool definitions
• src/client.js - Scraper initialization
• src/*.test.js - Unit tests

Troubleshooting

Server not appearing in Claude Desktop:

• Check that the path in config is absolute, not relative
• Check Claude Desktop logs for errors
• Ensure Node.js 18+ is installed

Scraper errors or timeouts:

• Welcome to the Jungle's site structure may have changed
• First request is slower as Chrome initializes
• Check your internet connection
• Consider reducing maxResults for faster responses

Chrome/Puppeteer issues:

• Puppeteer downloads Chromium automatically on first install
• If you see Chrome errors, try reinstalling: npm install puppeteer --force

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

MIT

Credits

• Built with Puppeteer for web scraping
• Built with the Model Context Protocol SDK
• Tested with Vitest

Disclaimer

This tool is for personal use only. Please respect Welcome to the Jungle's Terms of Service and robots.txt. Use responsibly and avoid excessive requests that could impact their servers.