Jobsuche MCP Server

An AI-friendly job search integration server using the Model Context Protocol (MCP). This server provides tools for searching German job listings via the Federal Employment Agency (Bundesagentur für Arbeit) API.

Features

• AI-Friendly Interface: Simple, semantic parameters for job searching
• Official API Integration: Uses the jobsuche crate for reliable API access
• Rich Filtering: Search by location, job title, employment type, contract type, and more
• Comprehensive Details: Get full job information including descriptions and requirements
• Pagination Support: Handle large result sets efficiently
• Zero Configuration: Works out of the box with sensible defaults

Installation

From npm (Recommended)

npm install -g @wunderfrucht/jobsuche-mcp-server

From Source

git clone https://github.com/wunderfrucht/jobsuche-mcp-server.git
cd jobsuche-mcp-server
cargo build --release

Configuration

The server uses environment variables for configuration (all optional):

• JOBSUCHE_API_URL: API base URL (default: official Bundesagentur für Arbeit API)
• JOBSUCHE_API_KEY: Custom API key (default: public API key)
• JOBSUCHE_DEFAULT_PAGE_SIZE: Default results per page (default: 25)
• JOBSUCHE_MAX_PAGE_SIZE: Maximum results per page (default: 100)

Usage with MCP Clients

Claude Desktop

Add to your Claude Desktop configuration (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):

{
  "mcpServers": {
    "jobsuche": {
      "command": "npx",
      "args": ["@wunderfrucht/jobsuche-mcp-server"]
    }
  }
}

Or using local binary:

{
  "mcpServers": {
    "jobsuche": {
      "command": "/path/to/jobsuche-mcp-server"
    }
  }
}

Continue.dev

Add to your Continue configuration:

{
  "mcpServers": {
    "jobsuche": {
      "command": "npx",
      "args": ["@wunderfrucht/jobsuche-mcp-server"]
    }
  }
}

Available Tools

1. search_jobs

Search for jobs in Germany using various filters.

Note: For most use cases, consider using search_jobs_with_details or batch_search_jobs instead, as they are more efficient for AI workflows.

Parameters:

• job_title (optional): Job title or keywords (e.g., "Software Engineer", "Data Scientist")
• location (optional): Location name (e.g., "Berlin", "München", "Deutschland")
• radius_km (optional): Search radius in kilometers from the location
• employment_type (optional): Employment type filter
  • Options: "fulltime", "parttime", "mini_job", "home_office", "shift"
• contract_type (optional): Contract type filter
  • Options: "permanent", "temporary"
• published_since_days (optional): Days since publication (0-100, default: 30)
• page_size (optional): Number of results per page (1-100)
• page (optional): Page number for pagination (starting from 1)
• employer (optional): Employer name to search for (e.g., "BARMER", "Siemens")
• branch (optional): Industry/branch to search in (e.g., "IT", "Gesundheitswesen")

Examples:

{
  "job_title": "Software Engineer",
  "location": "Berlin",
  "employment_type": ["fulltime"]
}

{
  "location": "München",
  "published_since_days": 7,
  "radius_km": 50
}

{
  "job_title": "Data Scientist",
  "location": "Deutschland",
  "employment_type": ["fulltime", "parttime"],
  "page_size": 50
}

{
  "employer": "BARMER",
  "location": "Wuppertal",
  "employment_type": ["parttime"]
}

2. get_job_details

Get detailed information about a specific job posting.

Parameters:

• reference_number (required): Job reference number from search results

Example:

{
  "reference_number": "10001-1234567890-S"
}

3. search_jobs_with_details ⭐ RECOMMENDED

Search for jobs and automatically fetch full details for top results in a single operation.

Why use this? Combines search_jobs + mult

... View full README on GitHub