agentleFS
Faaz17Faaz17/Agents-and-Skills/skills/research-writing/consultancy-register/SKILL.md

consultancy-register

Skill1 starsChanged 2 months agoMITView on GitHub

How to write the MGT4897 report so it reads as a professional consultant's report to an airport COO while still meeting academic marking standards — the executive voice, analytical rather than descriptive prose, recommendation structure, claim calibration, British English conventions, table and figure discipline, and the AI-tell patterns that must be edited out. Use when drafting or revising any prose, and as the style authority before submission.

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:

    Every paragraph should either advance an argument or be cut.

    line 29
  2. Distinctive:

    Never write "clearly", "obviously", or "it is well known that" — they assert rather than evidence.

    line 71
  3. Distinctive:

    Keep summary tables in the main text.

    line 115
  4. Distinctive:

    Read the full draft aloud.

    line 173

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

Register — consultant to a COO, marked by an academic

This report has two audiences at once, and the tension between them is the whole style problem.

The stated reader is the COO of Dubai International Airport. They want a decision, the evidence behind it, and the risks. They do not want a literature review.

The actual marker is an operations management academic. They want theory applied correctly, sources cited, methods justified, and limitations acknowledged.

Both are satisfied by the same thing: analysis that visibly drives a recommendation, with the reasoning and the sources on show. Neither is satisfied by description.

The core distinction: descriptive vs analytical

Descriptive (Pass)Analytical (Distinction)
"Forecasting is important for airport operations.""A one-million-passenger forecast error at DXB is roughly 2,700 passengers a day of misallocated immigration and baggage resource — recoverable within a season for staffing, unrecoverable for a terminal commissioned on the same number."
"Exponential smoothing was applied with alpha values of 0.2, 0.3 and 0.5.""Higher alpha values fit the post-2022 recovery better precisely because they discount the collapse faster — which is a reason to distrust them as a basis for a 2036 capacity decision, not a reason to prefer them."
"DWC scored highest in the weighted factor model.""DWC leads on land availability and long-run capacity, but its margin rests entirely on a 0.20 weighting for land constraint. At 0.10 the ranking reverses, so the recommendation is really a judgement that physical constraint at Al Garhoud outweighs near-term connectivity."
"TQM is a holistic quality approach.""DXB's quality exposure is peak-hour variance, not average performance, which points to statistical process control on queue times ahead of a programme-level TQM rollout."

Every paragraph should either advance an argument or be cut. At 2,500 words there is no room for a paragraph that only sets up another paragraph.

Paragraph architecture

Build each paragraph on this spine:

  1. Claim — the point, stated first, in one sentence
  2. Evidence — the computed figure or the cited source
  3. Analysis — what it means, why it holds, where it is contestable
  4. Implication — what the COO should do or conclude

The commonest failure is stopping after step 2. A number with no interpretation is a data point, not an argument. The second commonest is stopping after step 3 — analysis with no implication reads as academic exercise rather than consultancy.

Writing to a COO

  • Lead with the answer. Each section's first sentence states its conclusion. The reasoning follows. Do not build to a reveal
  • Quantify. "Demand exceeds design capacity by roughly 38 million passengers by 2035" beats "demand will significantly exceed capacity"
  • Name the decision. Recommendations should be actionable: what, by when, and what it depends on
  • Be honest about uncertainty. An executive discounts a report that hides its assumptions. State the basis, give a range where a range is honest
  • No throat-clearing. "This section will discuss…" wastes words the count cannot spare. Open with substance

Calibrating claims

Match language strength to evidence strength.

EvidenceLanguage
Figure given in the brief or a primary release"DXB handled 95.2 million passengers in 2025 (Dubai Airports, 2026)."
Computed from the provided data"The regression implies…", "On this basis the 2035 forecast is…"
Multiple independent sources agree"The evidence indicates…"
Single credible source"ICAO (2026) reports…"
Your inference"This suggests…", "A plausible reading is…"
Assumption"On the assumption that load factors hold…", "This model assumes…"

Never write "clearly", "obviously", or "it is well known that" — they assert rather than evidence. Never present a forecast as a fact. A forecast is a conditional statement, and the conditions belong in the sentence.

British English

Middlesex is a UK institution. Hold these throughout:

  • -ise not -ize: organise, analyse, prioritise, emphasise, utilise
  • -our: behaviour, labour, favourable
  • -re: centre, metre
  • programme (not program, except software), licence (noun) / license (verb), practice (noun) / practise (verb), enquiry, defence, fulfil, modelling, labelled, travelled
  • Dates: 4 September 2026
  • Currency: state the basis — AED or USD — and any conversion rate with its date
  • Numbers: "95.2 million" in prose, full precision in tables

The brief itself is written in American English ("analyses", "prioritise" mixed with "visualize", "Summarize"). Do not copy its spelling. Quote its wording where you quote it, and write British English everywhere else.

Structural conventions

  • Paragraph form for argument. Bullets only for genuine lists — the three operational recommendations, the location factors, a list of assumptions
  • Third person. "This report", "the analysis", "the recommended option" — not "I" or "we". The consultant framing tempts first person; resist it in the analysis. It is acceptable in a covering framing sentence if the module permits, but third person is safer
  • Present tense for what sources argue; past tense for what was done ("three methods were applied"); conditional for the projection ("demand would exceed capacity by…")
  • Define every acronym at first use: "Dubai World Central – Al Maktoum International Airport (DWC)", "mean absolute deviation (MAD)"
  • Numbered sub-sections (3.1, 3.2.1) — appropriate here and needed for appendix cross-referencing

Tables and figures

The brief requires visuals developed in MS Excel, and rubric item 8 marks their use.

  • Number and caption every one: "Table 3: Forecast accuracy comparison, DXB annual passengers, 2013–2025" · "Figure 2: Actual and forecast passenger traffic to 2036"
  • Refer to each in the text: "Table 3 shows…". A figure nobody refers to is decoration
  • State the source under each: "Source: author's calculations from Dubai Airports (2026)"
  • Keep summary tables in the main text; full workings go to the appendices
  • Table text counts toward the word count in Word — a further reason to keep workings in appendices

Editing out AI tells

These patterns read as machine-generated to markers and to Turnitin's AI indicator.

Vocabulary to cut: delve, leverage (as a verb), robust, pivotal, crucial, seamless, underscore, testament, landscape (figurative), realm, tapestry, myriad, plethora, navigate (figurative), foster, harness, unlock, transformative, game-changing, comprehensive, holistic, multifaceted, ever-evolving, rapidly evolving, strategic imperative, paramount.

Constructions to cut:

  • Rule of three — "efficient, scalable, and sustainable". Real analysis rarely arrives in perfect triples. Use one precise adjective, or two, and vary
  • Negative parallelism — "It is not just X, it is Y". Almost always padding
  • Superficial -ing analysis — "…, highlighting the importance of…", "…, reflecting the growing need for…". These append an unearned conclusion to a fact
  • Vague attribution — "experts suggest", "studies show", "it is widely believed". Name the source or delete the claim
  • Promotional register — "stands as a testament to", "plays a vital role in". This is a report to an operator, not a press release. The temptation is acute when writing about Dubai — resist it
  • Em-dash overuse — one per paragraph at most
  • Formulaic openers — "In today's fast-paced aviation industry", "In an era of"
  • Hollow closers — "Ultimately, success will depend on a range of factors"

Rhythm: vary sentence length deliberately. Uniform 20-to-25-word sentences are the strongest AI signal of all. Put a short sentence after a long one.

Section openings that work

Not "This section will analyse the location options", but:

  • "Three options can absorb the forecast demand, and only one of them is unconstrained by land."
  • "The choice of forecasting model turns on a single question: whether the 2020–2022 collapse belongs in the fitting window."
  • "DXB's design capacity is 115 million passengers. On every basis tested, demand reaches it before 2036."

Word discipline at 2,500 words

The budget is tight for eight tasks. When over:

  1. Cut throat-clearing openers
  2. Cut restatements of what was already said
  3. Move workings to the appendices
  4. Cut adjectives
  5. Cut framework description the reader can infer, keep framework application
  6. Cut examples that duplicate a point already evidenced

Never cut citations, the analysis sentence, or the recommendation.

Before submission

Read the full draft aloud. Every sentence that is hard to say is hard to read — this catches more than any checklist.

  • No paragraph stops at evidence without interpretation
  • Every section opens with its conclusion
  • No claim stronger than its source
  • No forecast presented as a fact
  • No banned vocabulary or construction
  • British English throughout, consistently
  • Sentence length genuinely varied
  • Every acronym defined at first use
  • Every table and figure numbered, captioned, sourced, and referred to in the text
  • Recommendations are specific enough to act on

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.