agentleFS
Faaz17Faaz17/Agents-and-Skills/skills/research-writing/excel-deliverables/SKILL.md

excel-deliverables

Skill1 starsChanged 2 months agoMITView on GitHub

Building the MS Excel workbook the MGT4897 brief requires — live-formula forecasting sheets, error-metric comparisons, the associative regression, the weighted factor model, and the capacity gap analysis, plus the charts that go into the report. Covers the Excel functions to use, how to present tables and figures in the report, and why formulas beat pasted values. Use when producing any table or figure for the report.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Put every parameter — alphas, weights, capacity, the effective-capacity ratio — in its own labelled cell and reference it absolutely.

    line 64
  2. Distinctive:

    Keep the palette restrained — two or three colours.

    line 85
  3. Distinctive:

    No 3D effects, no gradient fills, no chart junk

    line 87
  4. Distinctive:

    Add a numbered caption below: *Figure 3: Forecast passenger demand against DXB design capacity, 2026–2036.

    line 95

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Excel deliverables

The brief requires it: "Your report should include tables and/or figures developed using MS Excel to visualize data and support your arguments." Rubric item 8 marks the use of visuals. This is not optional decoration.

Build the workbook

py .claude/skills/excel-deliverables/scripts/build_workbook.py

Writes 02-Analysis/DXB-Calculations.xlsx with seven sheets:

SheetContents
1-DataThe provided series — the only hard-typed numbers in the file
2-PassengersSix forecasting methods, full error columns, CFE/MAD/MSE/MAPE/TS, accuracy comparison, 2026 forecast, two charts
3-MovementsThe same for flight movements
4-AssociativePassengers regressed on movements; SLOPE/INTERCEPT/RSQ/STEYX; scatter with fitted line; expected traffic at 500k/550k/600k
5-Long-TermProjections to 2036 on three fitting bases, with a comparison chart
6-Weighted-FactorLocation decision template with SUMPRODUCT scoring
7-Capacity-GapForecast demand against the 115m design capacity, with gap and utilisation columns

Options: --capacity, --long-to, --at, -o.

Open it in Excel and press F9 before using anything from it. openpyxl writes formulas but does not evaluate them; the cached values are empty until Excel calculates. A screenshot taken before recalculation will show zeros or blanks.

Why live formulas, not pasted values

  1. The brief says the visuals must be developed using Excel. A sheet of typed numbers is not development; a sheet of formulas is
  2. Change one input — an alpha, a weight, the capacity assumption — and everything downstream updates. During revision this saves hours and prevents the classic error of a report whose tables no longer agree with each other
  3. A marker who opens the workbook can follow the arithmetic. Auditability is credibility
  4. It removes transcription error, which is the most common source of a number in the prose that does not match the number in the appendix

The Excel functions this report needs

PurposeFunction
Moving average=AVERAGE(B2:B4)
Weighted moving average=$B$23*B4+$C$23*B3+$D$23*B2 — weights in named parameter cells, never typed inline
Exponential smoothing=C4+$alpha*(B4-C4)
Linear trend forecast=TREND(known_y, known_x, new_x) or =FORECAST.LINEAR(x, known_y, known_x)
Regression slope / intercept=SLOPE(y,x) · =INTERCEPT(y,x)
Correlation and fit=CORREL(x,y) · =RSQ(y,x)
Standard error of estimate=STEYX(y,x)
CFE=SUM(error_range)
MAD=AVERAGE(abs_error_range)
MSE=AVERAGE(squared_error_range)
MAPE=AVERAGE(ape_range)
Tracking signal=CFE_cell/MAD_cell
Weighted factor total=SUMPRODUCT(weights, scores)
Best model by a metric=INDEX(names, MATCH(MIN(metric_range), metric_range, 0))

Put every parameter — alphas, weights, capacity, the effective-capacity ratio — in its own labelled cell and reference it absolutely. Parameters buried inside formulas cannot be sensitivity-tested, and sensitivity testing is where the marks are.

Charts for the report

Four earn their place. More than that and the 2,500 words disappear under captions.

FigureTypeShows
Actual passenger traffic 2013–2025LineThe series and, unmistakably, the COVID break
Forecast accuracy by methodColumnMAPE or MAD side by side — makes the model choice visual
Passengers against flight movementsScatter + trendlineThe associative relationship, with r² displayed
Forecast demand vs 115m design capacity to 2036Line, capacity as a horizontal referenceThe gap and the crossing year — this is the money chart for the COO

Chart discipline:

  • Title every chart with what it shows, not "Chart 1"
  • Label both axes, with units
  • Show the trendline equation and R² on the scatter (Excel: Add Trendline → Display equation and R-squared)
  • Keep the palette restrained — two or three colours. Screenshots go into a black-and-white- printable report
  • No 3D effects, no gradient fills, no chart junk

Getting Excel output into the report

  1. Recalculate (F9)
  2. Copy the chart or table range
  3. Paste into Word as a picture — Paste Special → Picture (Enhanced Metafile). It will not reflow, break, or lose its formatting when the document is converted to PDF
  4. Add a numbered caption below: Figure 3: Forecast passenger demand against DXB design capacity, 2026–2036. Source: author's calculations from Dubai Airports (2026).
  5. Refer to it in the text: "Figure 3 shows…". An unreferenced figure is decoration

For tables that must remain editable text — the weighted factor model, the accuracy comparison — rebuild them as native Word tables so they are searchable and so the borders follow the document standard. Keep the numbers identical to the workbook.

What goes in the appendices

Main text: the summary comparison table, and the three or four figures above.

Appendices: full method-by-method workings, every error column, the complete long-term projection table on all three bases, the full weighted factor matrix with justifications, and any sensitivity runs. Appendices do not count toward the 2,500 words — use them.

Label appendices A, B, C and refer to each from the main text: "the full workings are at Appendix B".

The weighted factor sheet is a template

6-Weighted-Factor ships with placeholder weights and every score set to 5. Replace all of it. The factors listed are plausible starting points, not answers, and the justification column is empty on purpose. Weights that arrive unexamined from a template are exactly what the brief warns against when it says weightings "must reflect Dubai Airports' specific strategic priorities, not generic ones".

Before the workbook is used

  • Opened in Excel and recalculated — no #REF!, #DIV/0!, #VALUE! anywhere
  • Weights on the weighted factor sheet sum to exactly 1.00
  • Every placeholder score and justification replaced
  • Numbers in the report match the workbook exactly
  • Charts titled, axes labelled, sources stated
  • Assumptions labelled as assumptions on the sheet, not presented as data
  • Submit the workbook alongside the report if the module accepts supporting files — check the Turnitin link, since Turnitin usually accepts only one document

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.