Daily Calorie Tracker

You can track daily calories through natural language interactions with this MCP server. It records meals, provides daily and weekly summaries, lets you search foods by calories, and stores data persistently for cross-session use.

How to use

You interact with the server through MCP clients by describing meals, asking for summaries, or searching food calories. Typical workflows include logging a meal with a natural description, viewing today’s calorie total with a breakdown, generating a weekly trend report, and looking up calories for a specific food. Use natural phrases like “Log breakfast: oatmeal with banana and coffee” or “Show today’s calorie summary” to trigger the corresponding tool.

How to install

Prerequisites: ensure you have Node.js installed on your system. You will also use npm or npx to run the MCP server.

1. Install the MCP server package globally using npm.

npm install -g daily-calorie-tracker-mcp

2. Run the MCP server locally in development or start mode. The server can be executed via npm run dev for development or via a built distribution for production.

3. Start a production build and run the server from the built output when you are ready for deployment.

npm run build
npm start

Additional configuration and quick-start examples

The following configurations enable integration with MCP clients. Use the exact commands shown to start the server in different environments.

{
  "mcpServers": {
    "daily-calorie-tracker": {
      "command": "npx",
      "args": ["-y", "daily-calorie-tracker-mcp"]
    }
  }
}

If you prefer a local start with a built distribution, use this form:

{
  "mcpServers": {
    "daily-calorie-tracker": {
      "command": "node",
      "args": ["/path/to/daily-calorie-tracker/dist/index.js"]
    }
  }
}

API endpoints and supported actions

You can perform the following actions via the MCP tools exposed by the server. Each action is available as a callable function through your MCP client.

Storage and data access

Data is stored in an SQLite database to persist across sessions. The database location is available on macOS/Linux at ~/.daily-calorie-tracker/calories.db and on Windows at %USERPROFILE%\.daily-calorie-tracker\calories.db.

Food database and examples

The server includes a broad food database with common foods across categories such as proteins, carbohydrates, vegetables, fruits, beverages, dairy, meals, and snacks. Use the search tool to retrieve calorie information for specific foods.

Development

Development commands you will use during your setup and testing are shown here.

npm run dev

npm run build

npm start

Troubleshooting and notes

If you encounter issues, verify that the MCP server process starts without errors and that the local SQLite database path is writable. Ensure your MCP client is configured with the correct server name and command arguments as shown in the configuration examples.

Available tools

add_meal

Log a meal with a natural language description and category (breakfast, lunch, dinner, or snack).

get_daily_summary

Retrieve today’s calorie total with a breakdown by meal.

get_weekly_report

Generate a weekly report with averages, trends, and achievement metrics.

search_food

Query the food database for calorie information on a specific food item.