Adzuna Job Search MCP
MCP server for Adzuna Job Search API - search jobs, analyze salaries, and research employers across 12 countries
Adzuna Jobs MCP Server
A Model Context Protocol (MCP) server that provides AI assistants with access to the Adzuna Job Search API. Search for jobs, analyze salary data, and research employers across 12 countries.
Features
- Job Search - Search millions of job listings with filters for location, salary, job type, and more
- Salary Analysis - Get salary histograms, regional comparisons, and historical trends
- Company Research - Find top employers by hiring volume
- Multi-Country Support - Access job markets in 12 countries with local currency support
Supported Countries
| Code | Country | Currency |
|---|---|---|
gb | United Kingdom | GBP £ |
us | United States | USD $ |
de | Germany | EUR € |
fr | France | EUR € |
au | Australia | AUD $ |
nz | New Zealand | NZD $ |
ca | Canada | CAD $ |
in | India | INR ₹ |
pl | Poland | PLN zł |
br | Brazil | BRL R$ |
at | Austria | EUR € |
za | South Africa | ZAR R |
Available Tools
| Tool | Description |
|---|---|
search_jobs | Search for jobs with filters (keywords, location, salary, job type) |
get_categories | Get valid job category tags for a country |
get_salary_histogram | Get salary distribution data for job searches |
get_top_companies | Get top employers by number of open positions |
get_geodata | Get salary data broken down by geographic region |
get_salary_history | Get historical salary trends over time |
get_api_version | Get current Adzuna API version |
Prerequisites
- Python 3.10+
- Adzuna API credentials (free)
Getting Adzuna API Credentials
- Go to Adzuna Developer Portal
- Sign up for a free account
- Create a new application
- Copy your App ID and App Key
Installation
Option 1: Install from PyPI (Recommended)
pip install adzuna-mcp
Or use uvx for isolated execution:
uvx adzuna-mcp
Option 2: Install from Source
git clone https://github.com/folarinakinloye/adzuna-mcp.git
cd adzuna-mcp
pip install -e .
Option 3: Development Setup
git clone https://github.com/folarinakinloye/adzuna-mcp.git
cd adzuna-mcp
python3 -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
pip install -e ".[dev]"
Configure Environment Variables
Create a .env file with your Adzuna credentials:
ADZUNA_APP_ID=your_app_id_here
ADZUNA_APP_KEY=your_app_key_here
Usage
With Claude Desktop
Add to your Claude Desktop configuration file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"adzuna-jobs": {
"command": "/path/to/adzuna-mcp/venv/bin/python",
"args": ["/path/to/adzuna-mcp/server.py"],
"env": {
"ADZUNA_APP_ID": "your_app_id",
"ADZUNA_APP_KEY": "your_app_key"
}
}
}
}
Restart Claude Desktop after updating the configuration.
With Cursor
Add to your Cursor MCP settings (.cursor/mcp.json in your project or global config):
{
"mcpServers": {
"adzuna-jobs": {
"command": "/path/to/adzuna-mcp/venv/bin/python",
"args": ["/path/to/adzuna-mcp/server.py"],
"env": {
"ADZUNA_APP_ID": "your_app_id",
"ADZUNA_APP_KEY": "your_app_key"
}
}
}
}
With Other MCP Clients
Run the server directly:
# Activate virtual environment
source venv/bin/activate
# Run with stdio transport (default)
python server.py
# Or use FastMCP CLI
fastmcp run server.py:mcp
Development Mode
FastMCP provides a dev mode with an interactive inspector:
fastmcp dev server.py
This opens a browser-based UI to test your tools.
Example Prompts
Once connected, you can ask your AI assistant:
Job Search:
- "Find software engineer jobs in London paying over £60,000"
- "Search for remote Python developer positions in the US"
- "Show me data science jobs in Germany"
Salary Research:
- "What's the typical salary for a machine learning engineer in the UK?"
- "Compare software engineer salaries between London and Manchester"
- "How have data scientist salaries changed over the past year?"
Company Research:
- "Which companies are hiring the most software engineers in London?"
- "Show me the top employers for finance jobs in New York"
Tool Details
search_jobs
Search for jobs with comprehensive filtering:
Parameters:
- country (required): Country code (e.g., "gb", "us")
- keywords: Search terms (e.g., "python developer")
- location: City, region, or postal code
- page: Page number (starts at 1)
- results_per_page: Results per page (max 50)
- salary_min/salary_max: Annual salary filter
- full_time/part_time/contract/permanent: Job type filters
- category: Category tag from get_categories
- sort_by: "date", "salary", or "relevance"
- max_days_old: Maximum listing age in days
get_categories
Get valid category tags before searching:
Parameters:
- country (required): Country code
Returns category tags like "it-jobs", "engineering-jobs", "finance-jobs"
get_salary_histogram
Understand salary distribution:
Parameters:
- country (required): Country code
- keywords: Filter by job type
- location: Filter by location
- category: Filter by category
get_top_companies
Find major employers:
Parameters:
- country (required): Country code
- keywords: Filter by job type
- location: Filter by location
- category: Filter by category
get_geodata
Compare salaries across regions:
Parameters:
- country (required): Country code
- keywords: Filter by job type
- location: Focus on sub-regions
- category: Filter by category
get_salary_history
Analyze salary trends:
Parameters:
- country (required): Country code
- keywords: Filter by job type
- location: Filter by location
- category: Filter by category
- months: Months of history (default 12, max ~24)
Important Notes
- Salaries are annual amounts in the local currency of the selected country
- Call
get_categoriesfirst to get valid category tags for your country - Many jobs don't list salaries - salary filters will exclude these jobs
- Rate limits apply - the Adzuna API has usage limits on free tier
Project Structure
adzuna-mcp/
├── server.py # Main MCP server
├── pyproject.toml # Package configuration
├── requirements.txt # Python dependencies
├── .env.example # Environment template
├── .env # Your credentials (gitignored)
├── tests/ # Test suite
│ ├── __init__.py
│ └── test_server.py
├── .github/
│ └── workflows/
│ ├── ci.yml # CI pipeline
│ └── publish.yml # PyPI publishing
├── CONTRIBUTING.md # Contribution guide
├── LICENSE
├── .gitignore
└── README.md
Troubleshooting
Server not appearing in Claude Desktop
- Check the Python path is correct in your config
- Ensure the virtual environment has all dependencies installed
- Restart Claude Desktop completely (Cmd+Q / Alt+F4)
- Check Claude Desktop logs:
~/Library/Logs/Claude/mcp*.log
Authentication errors
- Verify your API credentials at Adzuna Developer Portal
- Check the
.envfile has correct values - Ensure environment variables are passed in the MCP config
No results returned
- Try broader search terms
- Check the country code is valid
- Remove salary filters (many jobs don't list salaries)
- Verify the category tag is valid for that country
Contributing
Contributions are welcome! Please see CONTRIBUTING.md for guidelines.
License
MIT License - see LICENSE for details.
Acknowledgments
Related Servers
NullBR MCP Server
A server for searching and retrieving movie and media resource information via the MCP protocol.
Jina AI Search
Perform semantic, image, and cross-modal searches using Jina AI's neural search capabilities.
NRT Search
A near real-time search server for indexing and querying documents, implemented in Java.
Yandex Search MCP Server
Perform real-time web searches using the Yandex Search API.
Ubersuggest
Perform AI-assisted SEO analysis using Neil Patel's Ubersuggest platform.
Gemini MCP
Integrate search grounded Gemini output into your workflow.
Google Search
Web search and webpage content extraction using the Google Custom Search API.
Aviationstack
An MCP server using the AviationStack API to fetch real-time flight data including airline flights, airport schedules, future flights and aircraft types.
Obsidian Omnisearch
Search your Obsidian vault using the Omnisearch plugin via a REST API.
Google PSE/CSE
A Model Context Protocol (MCP) server providing access to Google Programmable Search Engine (PSE) and Custom Search Engine (CSE).