Overview
What is Tibber MCP Server?
Tibber MCP Server is a Model Context Protocol (MCP) server for Tibber, a Norwegian power supplier. It enables AI agents to query energy prices and consumption data from the Tibber API via tools.
How to use Tibber MCP Server?
Install Python 3.12 and uv, clone the repository, set up a virtual environment, and install dependencies. Configure the host (e.g., Claude Desktop, Roo Code) with the uv command, the server script path, and your TIBBER_API_TOKEN environment variable. The server exposes two tools: get_consumption_data() and get_price_and_home_info().
Key features of Tibber MCP Server
- Provides current, today's, and tomorrow's energy price info
- Retrieves hourly consumption data for the last 30 days
- Returns home info: owner, address, subscription details
- Exposes two clear tools for LLM interaction
- Integrates with Claude Desktop and Roo Code
Use cases of Tibber MCP Server
- Analyze power consumption patterns and identify peak hours
- Find the hour with highest power usage yesterday
- Query power consumption at a specific time (e.g., yesterday at 7 AM)
- Get the current energy price
- List the cheapest hours of tomorrow
- Compare whether tomorrow's energy price is higher or lower
FAQ from Tibber MCP Server
What is this server for?
It allows an AI agent to interact with the Tibber API to query energy prices and consumption data.
How do I get a Tibber API token?
Get your token from the Tibber developer portal under access token settings.
What data can I query?
Hourly consumption data for the last 30 days, current/today/tomorrow prices, and home info (owner, address, subscription).
What are the runtime requirements?
Python 3.12 and the uv package manager. The server uses the pyTibber library and FastMCP framework.
How do I configure the server with Claude Desktop?
Add a JSON entry in your MCP server configuration with the command uv, arguments including the path to the cloned folder, and the environment variable TIBBER_API_TOKEN set to your token.
