mcp-rdcopilot-niagara
Agentic modules
Operate a Niagara installation in natural language.
Engineering: building the station and replicating a validated set on the next sites.
| The problem it solves | Taking over a site starts with reading someone else's work; querying a portfolio requires knowing the tree and where each history lives; and commissioning goes screen by screen, from an engineering workstation. |
|---|---|
| The work it does for you | This module opens a Niagara station to a conversational assistant installed on the user's workstation. It speaks MCP (Model Context Protocol), the standard socket through which an assistant connects to a line-of-business application: you do not program the assistant, you plug it in. You write your request in natural language; the assistant walks the station — tree, points, histories, alarms, field buses — and answers from that specific installation. Changes go through an identified operator's account and are traced in the station's audit log. |
| Benefits for you |
|
| Use cases |
|
| Platform | Niagara ≥ 4.13.x, Python ≥ 3.10 |
| Version | 0.60.2 |
| Price | on request |
| Module family | Operate an installation in natural language |
You understand an agent by watching it work
We give that demo live.



