thingsboard-mcp-server

MCP.Pizza Chef: AnyContext-ai

The Thingsboard MCP Server is an MCP server designed to expose Thingsboard IoT platform data as structured, real-time context to large language models (LLMs). By connecting to Thingsboard, it enables LLMs to access telemetry, device states, and other IoT data, enhancing AI-driven workflows and applications with live sensor and device information. The server supports easy setup with environment configuration, virtual environments, and dependency management, making it straightforward to deploy and integrate with MCP clients. This server is ideal for developers building AI copilots, agents, or automation tools that require dynamic IoT context from Thingsboard.

Use This MCP server To

Provide real-time IoT device data to LLMs for analysis Enable AI agents to monitor and control Thingsboard devices Integrate sensor telemetry into AI-driven decision workflows Feed live environmental data into conversational AI tools Automate IoT device management using LLM-powered commands

README

Thingsboard MCP Server

Setup environment using uv

Windows

# Install uv
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

# Create virtual environment
uv venv

# Activate virtual environment
.venv\Scripts\activate

Linux

# Install uv
curl -LsSf https://astral.sh/uv/install.sh | sh

# Create virtual environment
uv venv

# Activate virtual environment
source .venv/bin/activate

Add environment variables

Create .env file: cp .env.example .env

Add the environment variables to allow the MCP server to connect to Thingsboard.

Install dependencies

uv pip install -r pyproject.toml

Run server

uv run src/thingsboard.py

thingsboard-mcp-server FAQ

How do I set up the Thingsboard MCP Server environment?

Install 'uv' for your OS, create and activate a virtual environment, then install dependencies using 'uv pip install -r pyproject.toml'.

How do I configure the Thingsboard MCP Server to connect to my Thingsboard instance?

Create a .env file from the example and add your Thingsboard connection environment variables as specified.

How do I run the Thingsboard MCP Server?

After setup, run the server with the command 'uv run src/thingsboard.py' inside the activated virtual environment.

Can the Thingsboard MCP Server provide real-time telemetry data to LLMs?

Yes, it exposes live Thingsboard device telemetry and states as structured context for LLM consumption.

Is the Thingsboard MCP Server compatible with multiple LLM providers?

Yes, it is provider-agnostic and works with OpenAI, Anthropic Claude, Google Gemini, and other LLMs supporting MCP.

What programming language is used for the Thingsboard MCP Server?

The server is implemented in Python, leveraging standard virtual environment and package management tools.

How secure is the connection between the MCP server and Thingsboard?

Security depends on your Thingsboard instance configuration and environment variable management; the server uses standard secure connection practices.

Can I customize the data exposed by the Thingsboard MCP Server?

Yes, you can modify the server code or configuration to tailor the data and context exposed to the LLM.