thingsboard/thingsboard-mcp

Overview

The ThingsBoard MCP Server implements the Model Context Protocol to provide a natural language interface for large language models and AI agents. It enables them to interact with the ThingsBoard IoT platform by translating LLM requests into platform API calls.

Best for

Best for
Developers building AI assistants that need to query or manage a ThingsBoard IoT deployment

Use cases

• Letting an LLM query device telemetry data from ThingsBoard
• Allowing AI agents to manage IoT dashboards and alarms via natural language
• Enabling conversational interfaces for ThingsBoard platform administration

How to use

Tools exposed

• THINGSBOARD_URL
• THINGSBOARD_API_KEY
• THINGSBOARD_USERNAME
• THINGSBOARD_PASSWORD
• THINGSBOARD_LOGIN_INTERVAL_SECONDS
• HTTP_BIND_ADDRESS
• HTTP_BIND_PORT
• SPRING_AI_MCP_SERVER_STDIO
• SPRING_WEB_APPLICATION_TYPE
• SPRING_AI_MCP_SERVER_SSE_ENDPOINT
• SPRING_AI_MCP_SERVER_SSE_MESSAGE_ENDPOINT
• THINGSBOARD_CONNECTION_MAX_RETRIES
• THINGSBOARD_CONNECTION_RETRY_DELAY_SECONDS
• THINGSBOARD_CONNECTION_CONNECT_TIMEOUT_SECONDS
• THINGSBOARD_CONNECTION_READ_TIMEOUT_SECONDS
• LOG_LEVEL_APP
• LOG_LEVEL_TOOLS
• LOG_LEVEL_TOOL_RESPONSE
• LOGGING_PATTERN_CONSOLE
• LOGGING_CONSOLE_TARGET

Tested with

Claude Desktop, Cursor, VS Code Copilot, Claude Code

Example client config

{\n  "mcpServers": {\n    "thingsboard": {\n      "command": "docker",\n      "args": ["run", "-i", "--rm", "-e", "THINGSBOARD_URL", "-e", "THINGSBOARD_API_KEY", "thingsboard/mcp"],\n      "env": {\n        "THINGSBOARD_URL": "https://thingsboard.cloud",\n        "THINGSBOARD_API_KEY": "YOUR_API_KEY"\n      }\n    }\n  }\n}

Notes

The ThingsBoard MCP Server implements the Model Context Protocol to provide a natural language interface for large language models and AI agents. It enables them to interact with the ThingsBoard IoT platform by translating LLM requests into platform API calls.

96 stars on GitHub. Last updated 2026-03-16. Licensed Apache-2.0.

Use cases

• Letting an LLM query device telemetry data from ThingsBoard
• Allowing AI agents to manage IoT dashboards and alarms via natural language
• Enabling conversational interfaces for ThingsBoard platform administration

Pros

• Bridges LLMs directly to ThingsBoard without custom integration code
• Uses the open MCP standard, making it compatible with various LLM hosts
• Written in Java, aligning with ThingsBoard’s backend language

Cons

• Requires a running ThingsBoard instance to be useful
• Dependent on the MCP protocol ecosystem, which is still evolving
• Limited to functionality exposed via ThingsBoard’s REST API

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.