academic-register
Skill1 starsChanged 2 months agoMITView on GitHub
How to write MBA capstone prose that reads as distinction-level human academic writing — critical rather than descriptive voice, evidence-led paragraph construction, hedging and claim calibration, British English conventions, and the AI-tell patterns that must be edited out. Use when drafting or revising any prose for the proposal or report, and as the style authority before any submission.
In this repository37 other files it gives its agents.
- wsb-data-pipelineSkill
- docx-exportSkill
- harvard-mdxSkill
- humanise-docx-reportSkill
- make-pdfSkill
- bom-builderSkill
- ring-testSkill
- batch-draftSkill
- cold-emailSkill
- cv-intakeSkill
- cv-upgradeSkill
- find-contactSkill
- followupSkill
- humanize-emailSkill
- log-outreachSkill
- outreach-statusSkill
- target-researchSkill
- erc-statusSkill
- log-sessionSkill
- score-auditSkill
- submission-checkSkill
- sumo-statusSkill
- trial-runSkill
- vr-statusSkill
- aviation-data-sourcesSkill
- business-frameworksSkill
- consultancy-registerSkill
- excel-deliverablesSkill
- forecasting-methodsSkill
- operations-frameworksSkill
- secondary-data-sourcesSkill
- archive-indexSkill
- event-proposalSkill
- meeting-minutesSkill
- read-boxnoteSkill
- captureSkill
- new-scriptSkill
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
- Distinctive: line 20
Every paragraph should either advance an argument or be cut.
- Distinctive: line 46
Never write "clearly", "obviously", or "it is well known that" — they assert rather than evidence.
- Distinctive: line 108
Instead of "This section will discuss…", open with the substance:
- Distinctive: line 119
Read the full draft aloud, or have it read aloud.
≈ “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.
Files
Academic register — MBS4820
The core distinction: descriptive vs critical
Almost every mark lost on writing quality traces to one habit — describing what sources say instead of doing something with them.
| Descriptive (Pass) | Critical (Distinction) |
| "Porter (1980) identifies five forces." | "Porter's (1980) framework treats industry structure as stable, which understates how quickly platform entrants have reshaped last-mile economics." |
| "The market is growing rapidly." | "Growth of X% annually (Source, 2025) outpaces incumbent capacity expansion of Y% (Source, 2024), leaving an addressable gap of roughly Z parcels per day." |
| "Several authors have studied this." | "Autio and Zander (2016) attribute this to institutional voids; Sarasvathy (2001) locates it in founder logic instead. The distinction matters here because…" |
| "FedEx is a large logistics company." | "FedEx's scale advantage in intercontinental air freight does not transfer to intra-GCC ground density, where its cost per drop is structurally higher than a regional operator's." |
Every paragraph should either advance an argument or be cut.
Paragraph architecture
Build each paragraph on this spine:
- Claim — the point, stated first. One sentence.
- Evidence — cited data or literature supporting it.
- Analysis — what the evidence means, why it holds, where it is contestable.
- Link — how this bears on the venture or the next point.
The most common failure is a paragraph that stops after step 2. Evidence without interpretation is a quotation, not an argument.
Calibrating claims
Match the strength of the language to the strength of the evidence.
| Evidence | Language |
| Audited filing, official statistic | "FedEx reported…", "The data show…" |
| Multiple independent sources agree | "The evidence indicates…", "Consistently across sources…" |
| Single credible source | "Statista (2025) reports…", "One estimate places…" |
| Your inference from data | "This suggests…", "A plausible reading is…" |
| Projection or assumption | "On the assumption that…", "This model assumes…" |
Never write "clearly", "obviously", or "it is well known that" — they assert rather than evidence. Never present a projection as a finding.
British English
Middlesex is a UK institution. Hold these throughout:
-isenot-ize: organise, analyse, recognise, prioritise, emphasise-our: behaviour, labour, favourable-re: centre, metre, theatre- programme (not program, except software), licence (noun) / license (verb), practice (noun) / practise (verb), enquiry, defence, fulfil, modelling, labelled
- Dates: 18 August 2026
- Currency: state the basis — AED or USD — and any conversion rate with its date
- Single quotation marks for article titles in the reference list, as shown in
harvard-mdx
Structural conventions
- Paragraph form, not bullets. The module team says so explicitly. Bullets are acceptable only for genuine lists — SMART objectives, appendix items.
- Third person. Avoid "I" and "we" in the proposal and report body; use "this project", "the proposed venture", "the analysis". Some supervisors permit first person in reflective sections — ask before using it.
- Present tense for what sources argue ("Porter argues"), past tense for what was done ("data were collected"), future or conditional for the plan ("the venture will target").
- Define every acronym at first use: "courier, express and parcel (CEP)".
- Tables and figures numbered and captioned, each referred to in the text. A figure nobody refers to is decoration.
Editing out AI tells
These patterns read as machine-generated to markers and to Turnitin's AI indicator. Strip
them. (The humanizer skill covers this ground in depth; this is the capstone-specific
subset.)
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 changing.
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. State the conclusion properly or drop it.
- 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", "revolutionary". Academic writing is flat and precise.
- Em-dash overuse — one per paragraph at most; prefer commas, colons, and full stops.
- Formulaic openers — "In today's fast-paced business environment", "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. Let some paragraphs run five sentences and others two.
Section-opening patterns that work
Instead of "This section will discuss…", open with the substance:
- "The UAE's courier market divides into three tiers, and only one is contestable by a new entrant."
- "Two features of FedEx's cost structure matter for this venture."
- "The methodology is shaped by a single constraint: no primary data."
Signposting is fine in a 10,000-word report; announcing is not.
Before submission
Read the full draft aloud, or have it read aloud. Every sentence that is hard to say is hard to read. This catches more than any checklist.
Then verify:
- No paragraph stops at evidence without interpretation
- No claim stronger than its source
- No banned vocabulary or construction from the list above
- British English throughout, consistently
- Sentence length genuinely varied
- Every acronym defined at first use
- Every table and figure referenced in the text
- Paragraph form, third person
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
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.

