cobanov/teslamate-mcp

TeslaMate MCP Server

A Model Context Protocol (MCP) server that connects your TeslaMate database to AI assistants, enabling natural language queries about your Tesla data.

Features

• 🚗 18 Predefined Queries - Battery health, efficiency, charging, driving patterns, and more
• 🔍 Custom SQL Support - Execute safe SELECT queries with built-in validation
• 🗄️ Database Schema Access - Explore your TeslaMate database structure
• 🔒 Optional Authentication - Bearer token support for remote deployments
• 🏗️ Modular Architecture - Clean, maintainable codebase

Prerequisites

• TeslaMate running with PostgreSQL
• Python 3.11+ (for local) or Docker (for remote)

Quick Start

Local Setup (Cursor/Claude Desktop)

git clone https://github.com/cobanov/teslamate-mcp.git
cd teslamate-mcp
cp env.example .env
# Edit .env with your DATABASE_URL
uv sync

Configure your MCP client:

{
  "mcpServers": {
    "teslamate": {
      "command": "uv",
      "args": ["--directory", "/path/to/teslamate-mcp", "run", "main.py"]
    }
  }
}

Remote Setup (Docker)

git clone https://github.com/cobanov/teslamate-mcp.git
cd teslamate-mcp
cp env.example .env
# Edit .env with your DATABASE_URL
docker-compose up -d

Server available at: http://localhost:8888/mcp

Configuration

Create .env file:

DATABASE_URL=postgresql://user:pass@host:5432/teslamate
AUTH_TOKEN=                    # Optional: for remote auth

Generate auth token (optional):

python utils/generate_token.py

Available Tools

Predefined Queries (18 tools)

Vehicle Info:

• get_basic_car_information - VIN, model, firmware
• get_current_car_status - Real-time status, location, battery
• get_software_update_history - Firmware update timeline

Battery & Health:

• get_battery_health_summary - Current health metrics
• get_battery_degradation_over_time - Historical capacity
• get_daily_battery_usage_patterns - Usage patterns
• get_tire_pressure_weekly_trends - Tire pressure tracking

Driving Analytics:

• get_monthly_driving_summary - Monthly statistics
• get_daily_driving_patterns - Driving habits
• get_longest_drives_by_distance - Top trips
• get_total_distance_and_efficiency - Lifetime stats
• get_drive_summary_per_day - Daily summaries

Efficiency:

• get_efficiency_by_month_and_temperature - Seasonal analysis
• get_average_efficiency_by_temperature - Temperature impact
• get_unusual_power_consumption - Anomaly detection

Charging & Location:

• get_charging_by_location - Charging patterns
• get_all_charging_sessions_summary - Complete history
• get_most_visited_locations - Frequent places

Custom Queries (2 tools)

• get_database_schema - View database structure
• run_sql - Execute custom SELECT queries (read-only, validated)

Example Queries

"What's my current battery health?"
"Show me my longest drives"
"How does cold weather affect my efficiency?"
"Where do I charge most often?"
"Run a SQL query to find drives over 100km"

Project Structure

teslamate-mcp/
├── src/              # Core modules
│   ├── config.py     # Configuration
│   ├── database.py    # DB operations
│   ├── tools.py      # Tool registry
│   └── validators.py # SQL validation
├── queries/          # 18 SQL query files
├── data/             # Database schema
├── utils/            # Helper scripts
├── main.py           # Local (STDIO)
├── main_remote.py    # Remote (HTTP)
├── Dockerfile
└── docker-compose.yml

Development

Adding New Queries

1. Create SQL file in queries/:

-- queries/my_query.sql
SELECT * FROM my_table;

2. Add to src/tools.py:

ToolDefinition(
    name="get_my_data",
    description="What this returns",
    sql_file="my_query.sql",
)

3. Restart server - tool auto-registers!

Testing

python test_server.py

Security

• Authentication: Optional bearer token for remote access
• SQL Validation: Only SELECT queries allowed
• Read-only: No data modification possible
• Use HTTPS: In production environments

License

This project is licensed under the MIT License. See the file for details.

Acknowledgments

• TeslaMate - Tesla data logging
• Model Context Protocol - AI integration protocol