Agent examplesEnergy and commodities research

Supply and demand balance

Supply by train and demand by region on one unit, then the balance to 2030 in base, delayed and disruption cases. Every figure traced to a dated source

Customize this Agent

XLSXDOCX12 steps

The problem

New LNG capacity due by 2030, quoted in three units on two bases

Golden Pass is 18.1 million tonnes a year in its operator's release and 15.6 in the IEA's tracker: peak against baseload. Since March, QatarEnergy has said two damaged trains at Ras Laffan need three to five years to repair and North Field East could slip by more than a year. Count every train at full capacity from its announced start and the surplus is overstated

The request

Build our global LNG balance for 2026 to 2030 as a workbook and market note. Supply by train for every plant operating or past FID, as output not nameplate, timed and ramped as operators last announced, and Ras Laffan and North Field East as QatarEnergy last guided. Demand by region from the IEA and importers' plans, Europe on the EU's Russian LNG phase out. Million tonnes a year, conversion factors stated, baseload apart from peak. Cases by year: base; delayed, trains not yet producing slip twelve months, longer for known contractor, feedgas or sanctions problems; disruption, Hormuz restricted twelve more months. Flag where sources disagree and don't drop a project without asking. Show what moved since our June balance, attached.

LNG balance June 2026.xlsx

How Axe works

ApprovedChanges asked

Your reviewApprove, or say what to change

Approved or changes

Balance workbook and noteWorkbook, note and source check

Final reviewApprove the revised files

Revised balance workbook and noteRevised files and source check

Step

Build supply train by train

Prompt

Sent to the model at this step

Build the supply side one liquefaction train at a time, for every plant that is operating, commissioning, under construction or past FID. For each train record the country, project, operator, status, FID date, baseload and peak capacity exactly as the source states them and in the source's own unit, the first LNG date, the date that guidance was given, and the source: operator release, FID announcement, FERC or DOE filing, or the IEA's Global LNG Capacity Tracker. Take start dates from the operator's latest statement and note where an agency shows a different one. The balance needs output, not capacity: for each existing plant, compute its 2025 utilization as 2025 output against nameplate, from published trade data and operator reports, and apply it forward, adjusting only for disclosed outages, feedgas decline or damage. Give each new train a ramp up from the operator's guidance, or from the plant's earlier trains where there is none, then the utilization of comparable plants, and label both as assumptions. Carry the damaged Ras Laffan trains and North Field East as QatarEnergy last guided, with the range it gave. Give Qatar's undamaged trains and Das Island each a restart date and a Hormuz transit assumption: the operator's own stated date where there is one, otherwise the date in the Gulf assumptions, labelled as an assumption. Put sanctioned volume, such as Arctic LNG 2, on its own line, counted only against the buyers who can take it. Record any known contractor, feedgas or sanctions risk for each project not yet producing, with its source. List pre FID projects on their own tab, out of the base case unless I include them. Never drop a project; flag it with the reason.

Uses

Scope
As of date
Gulf assumptions

Produces

Train register

What you get back

XLSXBalance workbookSupply by train, demand by region, balance by case, changes since June and sources

DOCXMarket noteBalance by year, the projects that move it, changes since June and source disagreements

Source checkOpen items first, then each train and figure marked checked or corrected

Make it yours

Change the regions, the delay or disruption you test or the sources you rely on, or start from your own project list. Every step and every prompt can be edited

Customize this Agent

Opens a private copy of this Agent in your workspace, ready to adapt to your own sources

More agent examples

Power market analystsLevelized cost of energy analysisXLSXPPTX15 steps

Capital markets analystsBroker opinion of valuePPTXXLSX16 steps

Financial planning analystsBudget variance analysisXLSXDOCX13 steps

Credit research analystsCapital structure analysisXLSX12 steps

Life sciences analystsClinical trial landscapeXLSXDOCX14 steps