Overview
The ThingsBoard MCP Server provides a natural language interface enabling AI systems to access and manipulate data in ThingsBoard through the Model Context Protocol (MCP). With this server, you can ask conversational questions like querying devices by type, retrieving telemetry, and managing entities (devices, assets, customers, users, and entity groups) and their relationships, all in a structured response. It can simulate or insert time-series data, analyze IoT data to identify anomalies, spikes, or data gaps, and generate reports using AI tools. The MCP server supports querying and operating on entities, time-series, attributes, alarms, and relationships, and offers automation capabilities to streamline ThingsBoard operations via AI-powered workflows. It is designed to work with MCP clients such as Claude Desktop, Cursor, and other MCP protocol-compatible applications, enabling seamless integration with various AI assistants and LLMs. The server can run in STDIO or SSE modes, ships as Docker image or standalone JAR, and is configurable via environment variables to connect to the user\'s ThingsBoard instance and credentials.
Features
Entity Operations
Create, delete, view and manage devices, assets, customers, and users; manage entity groups; assign/unassign entities and handle related relationships.
Telemetry Management
Access attribute keys/values by scope, retrieve time-series data with aggregations, and insert/update telemetry with optional TTL settings.
Relations
Create, delete, discover, and navigate relationships between entities using direction-based queries.
Alarms
Create, delete, fetch alarms and alarm types, and retrieve severity information for specific entities.
Entity Data Query
Run complex queries over platform entities (devices, assets, customers, etc.) and retrieve data (fields, attributes, telemetry) in a structured, paginated format.
