2hr-jobs-mcp-server
OfficialClick on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@2hr-jobs-mcp-serverfind remote Python senior jobs above 20 000 PLN B2B"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
2hr-jobs-mcp-server
MCP server for Polish IT job board 2hr.pl — search IT jobs, salary data and market analytics via Model Context Protocol.
Built with FastMCP and Python 3.12. Works with Claude Desktop, Cursor, Windsurf, VS Code, and any MCP-compatible AI client.
What is MCP?
Model Context Protocol (MCP) is an open standard by Anthropic that allows AI models like Claude to securely call external APIs and databases in real time. Instead of pasting data manually into a chat, your AI agent fetches live data automatically.
This server connects Claude (or any MCP-compatible client) to the 2hr.pl job board — so you can ask:
"Find remote Python senior jobs above 20 000 PLN B2B"
"Is 18 000 PLN fair for a mid React developer in Warsaw?"
"Compare salaries: PHP vs Go for a senior in Poland"
"What skills do I need to become a DevOps engineer?"
"Which technologies are trending in Poland right now?"
Related MCP server: job-search-mcp
Tools
Job search
Tool | Description |
| Search IT jobs by keyword, city, salary range, experience level, contract type |
| Remote-only positions, optional keyword and salary filter |
| Jobs in a specific Polish city (Warsaw, Krakow, Wroclaw, Gdansk, Poznan...) |
| Newest job listings added to the board |
Salary analysis
Tool | Description |
| Full salary report for a technology: median, p25/p75/p90, B2B vs UoP, by city, trend |
| Side-by-side salary comparison of two technologies |
| Assess whether a job offer salary is fair vs current market data |
Market analytics
Tool | Description |
| Technology ranking by active job count with monthly trend |
| Compare two IT roles: jobs count, salary, required skills, remote availability |
| IT job market trends for 1m / 3m / 6m / 1y periods |
| Must-have, nice-to-have, and differentiating skills for a role |
| Companies with most active IT job listings, filterable by tech and city |
Quick start
Prerequisites
Python 3.10+
Redis (for caching)
A jobs REST API backend (configure in
.env)
Install
git clone https://github.com/your-username/2hr-jobs-mcp-server.git
cd 2hr-jobs-mcp-server
python3 -m venv venv
source venv/bin/activate # Linux/macOS
# .\venv\Scripts\activate # Windows
pip install -r requirements.txt
cp .env.example .env
# Edit .env — set JOBS_API_URL, JOBS_API_KEY, REDIS_URLRun locally (stdio transport)
python server.pyRun as HTTP server
python server_http.py
# Starts on http://0.0.0.0:8765Test with MCP Inspector
npx @modelcontextprotocol/inspector python server.py
# Opens http://localhost:5173 — call any tool interactivelyClaude Desktop integration
Edit the Claude Desktop config file:
macOS:
~/Library/Application Support/Claude/claude_desktop_config.jsonWindows:
%APPDATA%\Claude\claude_desktop_config.jsonLinux:
~/.config/claude/claude_desktop_config.json
{
"mcpServers": {
"2hr-jobs": {
"command": "/path/to/venv/bin/python",
"args": ["/path/to/2hr-jobs-mcp-server/server.py"],
"env": {
"JOBS_API_URL": "https://api.2hr.pl/v1",
"JOBS_API_KEY": "your_api_key",
"REDIS_URL": "redis://localhost:6379/0"
}
}
}
}Restart Claude Desktop. The server appears in the tools panel — Claude will automatically call the right tool based on your questions.
Cursor integration
Edit ~/.cursor/mcp.json (global) or .cursor/mcp.json (per project):
{
"mcpServers": {
"2hr-jobs": {
"command": "python",
"args": ["/path/to/server.py"],
"env": {
"JOBS_API_URL": "https://api.2hr.pl/v1",
"JOBS_API_KEY": "your_api_key"
}
}
}
}In Cursor's Agent Mode (Ctrl+Shift+I), the tools are available automatically.
Remote HTTP server (production)
For production use, configure the client with the remote URL:
{
"mcpServers": {
"2hr-jobs-remote": {
"url": "https://mcp.your-domain.com/sse",
"transport": "sse"
}
}
}Docker deployment
cp .env.example .env
# Fill in JOBS_API_URL and JOBS_API_KEY
docker compose up -d
# Check status
docker compose ps
docker compose logs mcp-serverFor HTTPS, use the included nginx.conf as a reverse proxy template with Let's Encrypt.
Project structure
2hr-jobs-mcp-server/
├── server.py # Entry point — stdio transport (Claude Desktop, Cursor)
├── server_http.py # Entry point — HTTP/SSE transport (production)
├── api/
│ └── client.py # Async httpx client for backend REST API
├── cache/
│ └── redis_cache.py # Redis cache with cache-aside pattern
├── tools/
│ ├── search.py # search_jobs, find_remote_jobs, find_jobs_in_city, get_latest_jobs
│ ├── salary.py # get_salary_report, compare_salaries, check_salary_fairness
│ └── analytics.py # get_top_technologies, compare_roles, get_market_trends, ...
├── resources/
│ └── documentation.py # MCP Resources (API docs, market summary)
├── prompts/
│ └── templates.py # MCP Prompts (job search assistant, salary negotiation)
├── security/
│ └── validators.py # Input sanitization, rate limiting, injection detection
├── monitoring/
│ └── metrics.py # Tool call tracking, latency monitoring
├── tests/
│ └── test_search.py # Unit tests for validators and formatters
├── Dockerfile
├── docker-compose.yml
├── nginx.conf
└── .env.exampleConfiguration
All settings via environment variables (copy .env.example to .env):
Variable | Default | Description |
| — | Backend REST API base URL |
| — | API authentication key |
|
| Redis connection string |
|
| Job listings cache TTL in seconds |
|
| Salary reports cache TTL |
|
| Analytics cache TTL |
|
| Market trends cache TTL |
|
| HTTP server port |
|
| Logging verbosity |
Running tests
pip install pytest pytest-asyncio
pytest tests/ -vSecurity features
Input sanitization: SQL injection, XSS, path traversal detection via regex patterns
Pydantic v2 validation with strict bounds on salary and limit parameters
Redis-based rate limiting per client identifier
Non-privileged
mcpuser in Docker containerAPI key authentication between MCP server and backend API
fail openon Redis errors — cache miss, not service failure
Architecture
AI Client (Claude / Cursor)
|
MCP Protocol
(stdio or HTTP/SSE)
|
MCP Server (Python)
FastMCP + tools/salary/analytics
|
Redis Cache (TTL 5-60 min)
|
REST API (your backend)
|
Database (MySQL / PostgreSQL)The MCP server is a stateless middleware layer — it holds no data of its own. All job and salary data lives in your existing backend, which the MCP server proxies through standardized tools.
Data source
All data is powered by 2hr.pl — Polish IT job board aggregating thousands of IT job listings.
Salaries in PLN gross/month
Covers all major Polish cities: Warsaw, Krakow, Wroclaw, Gdansk, Gdynia, Poznan, Lodz, Katowice, Rzeszow, Szczecin
Includes remote positions
Data refreshed continuously
Read the full tutorial (Polish): Jak zbudować serwer MCP w Pythonie dla portalu pracy
Related links
2hr.pl — Polish IT job board (data source)
Model Context Protocol spec — MCP specification by Anthropic
FastMCP — High-level Python MCP SDK
MCP Inspector — Interactive testing tool
Claude Desktop — MCP-compatible AI client
License
MIT — see LICENSE
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Latest Blog Posts
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/2hr-pl/2hr-jobs-mcp-server'
If you have feedback or need assistance with the MCP directory API, please join our Discord server