For DB
A lightweight, extensible MCP server for executing schema-aware SQL queries. Built on FastMCP and compatible with any SQLAlchemy-supported database.
Ask AI about For DB
Powered by Claude Β· Grounded in docs
I know everything about For DB. Ask me about installation, configuration, usage, or troubleshooting.
0/500
Reviews
Documentation
ποΈ MCP Server for Databases
A lightweight implementation of the Model Context Protocol (MCP) for database operations. This server enables AI models to interact with SQL databases through a structured, tool-based interface, providing safe and efficient database access.
π Table of Contents
- π Key Capabilities
- π§ Getting Started
- π§ MCP Interface
- π€ MCP Client Integration
- π§© LLM Integration
- β Error Handling
- π οΈ IDE and Tool Integration
- βοΈ Development
- π€ Contributing
- π License
- βοΈ MCP Server Configuration Reference
- π Acknowledgments
π Key Capabilities
π οΈ MCP Tool Integration
- Exposes database operations as MCP tools for AI model interaction
- Supports natural language prompts for database exploration
- Provides schema-aware query building assistance
πΎ Database Support
- Works with any SQLAlchemy-compatible database (SQLite, PostgreSQL, MySQL, etc.)
- Automatic schema inspection and reflection
- Transaction support for multi-statement operations
π‘ MCP Features
- Native MCP tools, resources, and prompts
- Support for multiple transport protocols (http, WebSocket, stdio)
- Structured error handling and validation
π§ Getting Started
β Prerequisites
- Python 3.8+
- SQLite or any other SQLAlchemy-supported database
βοΈ Installation
git clone https://github.com/devsenweb/mcp-server-for-db.git
cd mcp-server-for-db
python -m venv .venv
source .venv/bin/activate # macOS/Linux
# OR
.venv\Scripts\activate # Windows
pip install -r requirements.txt
π οΈ Configuration
Edit config.yaml:
db:
uri: sqlite:///./example.db # Supports any SQLAlchemy URI
(Optional) Create a sample DB:
python create_database.py
βΆοΈ Starting the MCP Server
python -m mcp_server.server --transport http --port 8080
# or
python -m mcp_server.server --transport ws --port 8080
# or
python -m mcp_server.server --transport stdio
π§ MCP Interface
π Tool Definitions
execute_query
{
"jsonrpc": "2.0",
"method": "tools/execute_query",
"params": {
"sql": "SELECT * FROM users WHERE age > 25"
},
"id": 1
}
validate_sql
{
"jsonrpc": "2.0",
"method": "tools/validate_sql",
"params": {
"sql": "SELECT * FROM users"
},
"id": 2
}
π Resource Endpoints
db://schema
{
"jsonrpc": "2.0",
"method": "resources/get",
"params": {
"uri": "db://schema"
},
"id": 3
}
db://tables
{
"jsonrpc": "2.0",
"method": "resources/get",
"params": {
"uri": "db://tables"
},
"id": 4
}
π οΈ IDE and Tool Integration
(Cursor IDE) .cursor/mcp.json
{
"mcpServers": {
"database-query": {
"url": "http://localhost:8000/mcp",
"transport": "http"
}
}
}
MCP Inspector
npm install -g @modelcontextprotocol/inspector
mcp-inspector --transport http --url http://localhost:8000/mcp
# Visit http://localhost:3000
βοΈ Development
pip install -r requirements-dev.txt
pytest tests/
π Project Structure
mcp-server-for-db/
βββ mcp_server/
β βββ __init__.py
β βββ server.py
β βββ db_adapter.py
βββ config.yaml
βββ create_database.py
βββ requirements.txt
π€ Contributing
Contributions are welcome! Feel free to submit a Pull Request.
π License
Licensed under the MIT License. See LICENSE for details.
βοΈ MCP Server Configuration Reference
python -m mcp_server.server --host 0.0.0.0 --port 8000 --transport http --config ./config.yaml
π Acknowledgments
- Powered by SQLAlchemy
- Based on the Model Context Protocol (MCP) standard
β οΈ Disclaimer
This project is a demo implementation of an MCP-compatible server for database interaction. It does not include SQL safety mechanisms, input sanitization, authentication, or permission controls.
β οΈ Use at your own risk. This server executes raw SQL and is intended for experimentation, prototyping, and local development only.
It is the responsibility of the user to implement proper guardrails, validation, and security measures before using this in any production or sensitive environment.
