Enapter MCP Server

A Model Context Protocol (MCP) server that provides seamless integration with the Enapter EMS. This server enables AI assistants and other MCP clients to interact with Enapter sites, devices and telemetry data.

Connecting Your AI Application

The Enapter MCP Server is available as a public hosted service at https://mcp.enapter.com/mcp. It uses streamable HTTP transport and OAuth 2.0 for authentication.

For specific instructions on how to connect your preferred AI client, please refer to the following guides:

• Claude Desktop

• Claude Mobile

• Claude Code

• Copilot Studio

• ChatGPT

Self-Hosting

If you prefer to run your own instance, you can self-host the server using Docker:

docker run --rm --name enapter-mcp-server \
  -p 8000:8000 \
  enapter/mcp-server:v1.4.0

The server can be configured using environment variables and command-line arguments.

Available Tools

The server exposes the following tools for interacting with the Enapter EMS:

Usage Examples

Here are realistic examples of how you can interact with your Enapter devices using AI assistants:

Example 1: Diagnostic Troubleshooting

User prompt:

One of our inverters just went offline. Can you check its status and tell me what the active alerts mean?

What happens:

• Server finds the specific inverter device

• Retrieves its current connectivity status and active alerts

• Reads the device blueprint to translate alert codes into human-readable descriptions

• Presents a summary of the issue to the user

Example 2: Historical Analysis & Performance

User prompt:

What was the average power consumption and temperature for the main HVAC system over the last 7 days?

What happens:

• Server locates the HVAC system device

• Checks its blueprint to identify the correct telemetry metric names for power consumption and temperature

• Fetches the historical telemetry data for the requested time period

• Calculates and presents the averages to the user

Example 3: Auditing Command Executions

User prompt:

Check if anyone tried to turn on the water pump this morning. Were there any errors during the execution?

What happens:

• Server locates the specific water pump device

• Reads the device blueprint to find the exact command name for "turning on" the device

• Searches the command execution history for that specific command executed this morning

• Retrieves the execution status and any associated error messages

• Reports back whether the command succeeded or failed

Example 4: Auditing Automation Rules

User prompt:

Can you check the automation rules running at the Alpha site? I need to verify the logic that automatically starts the electrolyser when excess solar power is available.

What happens:

• Server searches for rules within the specified site using search_rules

• Identifies the relevant rule based on its name/slug

• Retrieves the rule's Lua script using read_rule

• Analyzes the logic and confirms the exact threshold and conditions that trigger the electrolyser

Support

For issues, questions, or contributions, please:

• Email us at support@enapter.com.

• Open a GitHub Issue.

• Join our Discord server.

• Check our Contributing Guide.

Privacy Policy

For information about how we handle data, please refer to the Enapter Privacy Policy.

Made with ❤️ by Enapter