Lexiguard HR
Enables Cursor AI to query a local MySQL HR database in natural language using read-only SQL tools.
README
Lexiguard HR — Local MySQL MCP Server for Cursor
A small Model Context Protocol (MCP) demo that lets Cursor query a local MySQL HR database in natural language—without pasting credentials or schema into every chat.
Cursor (AI) ←── MCP (stdio) ──→ server.py ←── SQL ──→ MySQL (lexiguard)
What it does
| MCP tool | Description |
|---|---|
get_db_schema |
Lists tables, columns, types, and keys so the AI knows what exists |
query_hr_db |
Runs read-only SQL (SELECT, SHOW, DESCRIBE, EXPLAIN, WITH) and returns formatted text |
Example questions in Cursor chat:
- “Use get_db_schema to show all tables.”
- “Who works in Engineering?”
- “List employees with their department names.”
- “Average salary by department.”
Tech stack
- Python 3.10+
- MCP Python SDK (
mcp, FastMCP) - mysql-connector-python
- Cursor with project MCP config (
.cursor/mcp.json)
Prerequisites
Quick start
1. Create database and sample data
In MySQL Workbench or the CLI:
CREATE DATABASE IF NOT EXISTS lexiguard;
USE lexiguard;
CREATE TABLE IF NOT EXISTS departments (
id INT AUTO_INCREMENT PRIMARY KEY,
name VARCHAR(100) NOT NULL,
location VARCHAR(100)
);
CREATE TABLE IF NOT EXISTS employees (
id INT AUTO_INCREMENT PRIMARY KEY,
first_name VARCHAR(50) NOT NULL,
last_name VARCHAR(50) NOT NULL,
email VARCHAR(100) UNIQUE,
department_id INT,
hire_date DATE,
salary DECIMAL(10, 2),
FOREIGN KEY (department_id) REFERENCES departments(id)
);
INSERT INTO departments (name, location) VALUES
('Engineering', 'London'),
('HR', 'Manchester'),
('Finance', 'Birmingham');
INSERT INTO employees (first_name, last_name, email, department_id, hire_date, salary) VALUES
('Alice', 'Smith', 'alice.smith@lexiguard.com', 1, '2022-03-15', 65000.00),
('Bob', 'Jones', 'bob.jones@lexiguard.com', 1, '2021-07-01', 72000.00),
('Carol', 'Brown', 'carol.brown@lexiguard.com', 2, '2020-11-20', 48000.00),
('David', 'Taylor', 'david.taylor@lexiguard.com', 3, '2023-01-10', 55000.00);
2. Install Python dependencies
Windows (PowerShell):
cd db-mcp-server
python -m venv .venv
.\.venv\Scripts\pip install -r requirements.txt
macOS / Linux:
cd db-mcp-server
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
3. Configure environment
cp .env.example .env
Edit .env with your MySQL settings (especially MYSQL_PASSWORD if root has a password).
4. Connect Cursor
This repo includes project-level MCP config at .cursor/mcp.json.
- Open this folder as a Cursor workspace.
- Go to Cursor Settings → Models → MCP.
- Enable
lexiguard-hr(should show 2 tools). - Restart Cursor if the server does not appear.
macOS / Linux: Update .cursor/mcp.json so command points to your venv Python, for example:
"command": "${workspaceFolder}/.venv/bin/python"
Windows (default in this repo):
"command": "${workspaceFolder}/.venv/Scripts/python.exe"
Credentials are loaded from .env via envFile—do not commit .env (it is in .gitignore).
5. Try it
Open chat (Ctrl+L / Cmd+L), use Agent mode, and allow MCP tool runs when prompted.
Project layout
db-mcp-server/
├── .cursor/
│ └── mcp.json # Cursor MCP launch config
├── server.py # MCP server (FastMCP + MySQL)
├── requirements.txt
├── .env.example # Copy to .env locally
├── .gitignore
├── README.md # This file (demo / setup)
└── MCP-HOW-IT-WORKS.md # Short technical notes on MCP
Security notes
- Read-only queries only in
query_hr_db(noINSERT/UPDATE/DELETE). - Keep
.envlocal; use.env.exampleas a template for collaborators. - Intended for local development demos, not production exposure of a database.
Troubleshooting
| Issue | What to check |
|---|---|
| MCP server red / disconnected | MySQL running; .env password correct; re-run pip install |
| “Connection refused” | Start MySQL service; confirm MYSQL_HOST / MYSQL_PORT |
| Tools not listed | Reopen project folder; toggle MCP server; restart Cursor |
| GitKraken MCP error | Unrelated third-party config in global ~/.cursor/mcp.json |
Learn more
See MCP-HOW-IT-WORKS.md for a concise technical explanation of MCP and how this repo implements it.
License
Use and adapt freely for learning and demos.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.