MCP App Store
by KeenEthics
D3 Cloud icon

D3 Cloud MCP Server for ChatGPT

By D3 IoT

Overview

D3 Cloud monitors energy, gas, water and heat across commercial and residential buildings. Connect it to ask about your own estate in plain language: which sites you have, what each meter is using, how consumption compares month to month, the half-hourly load shape behind a bill, and which meters have stopped reporting. If you use D3 Bureau Services it also covers the commercial side: your supply contracts and rates, the invoices received against them, and the validation checks run on each one, including what passed, what failed and the difference identified. If you run tenant billing through D3, client administrators can ask about invoices, contracts, payments received and arrears, and get a time-limited link to download an invoice PDF. Payment details are never returned: the connector can say that a direct debit or card is on file, and nothing about the card or bank account behind it. You sign in with your own D3 Cloud account, and the connector sees exactly what that login sees and nothing wider. Access is re-resolved on every request, so a connection cannot outlast or exceed the permissions of the person who created it. It is read-only throughout: it can answer questions about your data but can never change it. Results carry their own caveats, so estimated days, part months and meters that were not counted are reported alongside the figures rather than quietly folded into them.

How to connect

  1. ChatGPT: Open the D3 Cloud listing in the ChatGPT app directory and choose Connect. ChatGPT listing D3 Cloud exposes 14 tools here.

Tools

consumption_by_meter

ChatGPT
Every meter at a site ranked by consumption, month by month. Returns columns: meterId, meter, location, total, unit, isMain.

consumption_by_utility

ChatGPT
Total consumption per utility for one or more sites — electricity, gas, water and heat side by side, each in its own unit. Electricity and heat are in kWh and water and gas in their own metered units, so the figures share no common scale and no total across them is meaningful. Returns columns: site, utility, total, unit, meters, months, partMonths.

consumption_daily_series

ChatGPT
A day-by-day series, reporting how many days were estimated rather than really read. Returns columns: day, usage, estimated.

consumption_hourly_profile

ChatGPT
Consumption hour by hour for one meter, in UK local time. Each value is the kWh used during that hour, which is numerically the same as the average kW demand across it — so an hourly figure answers a question asked in kW without conversion. Useful for overnight baseload, load shape and weekday-versus-weekend comparison. Built from settled readings, so it covers any period with readings rather than only recent days. Recorded for most electricity meters and some heat meters, and not for gas or water. For finer detail than an hour, use the interval series, which buckets device readings to 30 minutes over a shorter window. Returns columns: hour, usage, exported.

consumption_interval_series

ChatGPT
Consumption in 30-minute or hourly intervals for one meter, built from the readings the device sent. Use it for load shape, overnight baseload, out-of-hours running and for building your own aggregations. Thirty minutes is the finest interval available — raw individual transmissions are not served, because bucketed intervals survive a missed transmission and raw readings are cumulative registers rather than consumption. Returns columns: intervalStart, consumption, register, readings, coversMinutes.

consumption_monthly_by_site

ChatGPT
Closed monthly totals per site, counted from supply meters so submeters are not added on top. Returns columns: site, period, utility, total, unit, estimatedMonths.

consumption_register_readings

ChatGPT
The readings recorded for a meter: the cumulative register, and the consumption booked for the period ending on each date. This is what an invoice is reconciled against. The consumption column is already the difference between consecutive registers rather than a running total. Returns columns: readingDate, register, periodFrom, consumption, source, newBaseline, registerMovement, agrees.

consumption_site_ranking

ChatGPT
Highest and lowest consuming sites over a period. The basis for a "top / bottom sites" widget. Returns columns: site, total, unit.

d3_capabilities

ChatGPT
Lists every capability D3 Cloud offers, how many of them this connection can currently use, and which access it was not granted. The full catalogue is wider than any single login reaches, so a capability missing from this connection may still exist on the platform behind a permission or a scope.

estate_clients

ChatGPT
The clients this login can see, with how many sites and meters each has. In D3 Cloud a client owns sites and a site contains meters, so a client name identifies an organisation rather than a building. Returns columns: client, sites, meters.

estate_meter_health

ChatGPT
Consumption meters and the day each was last heard from. The basis for an "offline" widget. Returns columns: meterId, meter, site, lastReadingDay, daysSince, offline.

estate_meter_sources

ChatGPT
The collector behind each meter — LoRa, M-Bus, Modbus IoT and so on — with its device identifier and when that collector last heard from it. Use it to diagnose a meter that has stopped reporting, or to see what a site is instrumented with. Returns columns: meterId, meter, site, utility, collector, device, rawValueStore, commissioned.

estate_meters

ChatGPT
Meters at one or more sites, with what each measures and how it is read. Returns columns: meterId, meter, site, utility, unit, hardware, serial, mpan, location.

estate_sites

ChatGPT
Every site the viewer can see, and which client owns each one. Sites belong to a client; meters belong to a site. Use the client columns to answer "which buildings belong to X". Returns columns: site, client.

Example Prompts

Click any prompt to copy it.

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/d3-cloud"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/d3-cloud)