About
The LG Therma V MCP Server exposes a Modbus‑connected heat pump to the Model Context Protocol, enabling developers and AI tools to query and control LG Therma V units programmatically for advanced automation.
Capabilities

The LG Therma V MCP Server bridges the gap between modern AI assistants and legacy HVAC hardware by exposing a Modbus‑based heat pump as an MCP resource. For developers building intelligent home automation, this server solves a common pain point: accessing real‑time sensor data and control commands from an LG Therma V unit without writing custom drivers or dealing with proprietary protocols. By translating Modbus registers into MCP resources, the server lets an assistant like Claude read temperatures, fan speeds, and operating modes, or send set‑point changes—all through natural language.
At its core, the server implements the MCP specification to offer a consistent API. It registers a set of tools that represent common heat‑pump operations (e.g., “Set target temperature”, “Read current mode”). These tools are automatically discoverable by any MCP‑compliant client, enabling a conversational AI to list available actions and invoke them with minimal friction. The server also exposes prompt templates that guide the assistant on how to phrase responses, ensuring that users receive clear, context‑aware explanations or confirmations of their commands.
Key capabilities include:
- Modbus TCP connectivity: The server can be pointed at any Modbus‑over‑TCP gateway, such as a Waveshare RS485‑to‑Ethernet adapter, making it deployable on standard networked hardware.
- Dynamic resource discovery: The MCP client can query the server for available registers and tools, allowing the assistant to adapt to different firmware versions or custom configurations.
- Bidirectional control: Not only can the assistant read sensor values, but it can also write to registers, enabling full automation (e.g., turning the unit on/off or adjusting fan speed based on weather forecasts).
Typical use cases span from simple “set the thermostat to 72 °F” commands to sophisticated scenarios where an AI assistant orchestrates a multi‑step routine: “If the outdoor temperature drops below 50 °F, increase indoor setpoint by 2 degrees and log the change.” In smart‑home ecosystems, this server allows integration with voice assistants, mobile apps, or rule engines without reinventing Modbus communication layers.
What sets the LG Therma V MCP Server apart is its developer‑friendly design. The server ships with a ready‑made configuration file for Claude Desktop, and its API surface mirrors the standard MCP layout, so any existing MCP client can consume it out of the box. Because it is open‑source under MIT, developers can audit, extend, or adapt the protocol mapping to fit custom hardware. In short, this server turns a legacy heat pump into a first‑class citizen in the AI‑powered automation landscape, giving developers a powerful, low‑friction bridge between natural language commands and physical HVAC control.
Related Servers
MindsDB MCP Server
Unified AI-driven data query across all sources
Homebrew Legacy Server
Legacy Homebrew repository split into core formulae and package manager
Daytona
Secure, elastic sandbox infrastructure for AI code execution
SafeLine WAF Server
Secure your web apps with a self‑hosted reverse‑proxy firewall
mediar-ai/screenpipe
MCP Server: mediar-ai/screenpipe
Skyvern
MCP Server: Skyvern
Weekly Views
Server Health
Information
Explore More Servers
GUARDRAIL: Security Framework for Large Language Model Applications
Layered security for LLM and autonomous agent systems
Kaltura Events MCP
Manage Kaltura virtual events with AI-powered tools
Redis MCP Server
Fast in-memory MCP server powered by Redis
SAP-Claude MCP Server
Bridge Claude AI to SAP via Model Context Protocol
Miro MCP Server
AI agents control Miro boards via MCP
Entity Identification Server
Determine if two data sets belong to the same entity