OpenDSS MCP Server

MCP serverCommunity

Runs OpenDSS distribution studies from chat: IEEE test feeders, power flow, hosting capacity, DER siting and plots. By ahmedelshazly27.

4 starsUpdated 1 year agoGitHub data from 5 Nov 2026

At a glance

Type: MCP server
Works with: OpenDSS
How it connects: Drives a running copy of the tool
Access: Can modify models
Tool versions: OpenDSS 9.8 or later (README badge); Python 3.10 or later
Operating systems: Windows, macOS, Linux
Needs a OpenDSS licence: No
Transport: Local (stdio)
AI clients documented: Claude Desktop
Licence: MIT
Language: Python

What it can do

Loads the IEEE 13, 34 and 123 bus test feeders and runs snapshot, daily and yearly power flow with loss and voltage-profile results.
Checks voltages against ANSI C84.1, searches for DER placement and estimates hosting capacity.
Version 1.1 adds 13 tools: network diagnostics, batch scenario runs, feeder templates, multi-file DSS projects, circuit comparison and CSV/JSON/HTML export, for 20 tools in total.

Limitations

Last updated in October 2025; the PyPI package is still at 1.0.0 while the README describes v1.1.
The README's time-saving figures are the author's own claims.

Install

git clone https://github.com/ahmedelshazly27/opendss-mcp-server.git
cd opendss-mcp-server
pip install -e .

Source: github.com/ahmedelshazly27/opendss-mcp-server#installation

{
  "mcpServers": {
    "opendss": {
      "command": "python",
      "args": ["-m", "opendss_mcp.server"],
      "env": {
        "PYTHONPATH": "/absolute/path/to/opendss-mcp-server/src"
      }
    }
  }
}

Source: github.com/ahmedelshazly27/opendss-mcp-server#claude-desktop-configuration

MCP servers and skills run code on your computer. Review the source before installing.

AI-generated models, drawings and results must be checked by a qualified engineer before use.