academic

name	academic
description	Write a formal academic research report — Abstract, Background, Method, Findings, Discussion, References — with a selectable citation style (author-date/APA or numbered/Vancouver-IMRaD) and optional APA Method sub-sections. Use when the deliverable is a scholarly write-up for a research or technical-expert audience that demands traceable evidence and explicit method. Anti-trigger; for an informal trade-off narrative use google-design-doc, for a practitioner decision report built around a mandatory comparison table use engineering.
argument-hint	<the research question or topic to report on>

academic

Produces a formal academic research report: a scholarly write-up for a research or technical-expert audience — peers who will scrutinize method, evidence, and the limits of each claim before accepting it. Its center of gravity is the IMRaD structure plus mandatory citation discipline — the report is not conformant if any claim goes uncited, or if the chosen citation style is applied inconsistently. This genre follows the general IMRaD (Introduction/Background, Methods, Results, and Discussion) scholarly-paper convention, with a selectable citation style.

Pattern (industry: IMRaD scholarly paper, selectable citation style)

1. Abstract — 150-250 words: question, method, principal findings, conclusion.
2. Background / Related Context — what is already established and the gap this report addresses.
3. Method — how the reported evidence was gathered and verified; state the verification approach and how contested or weakened evidence was handled. In APA mode (optional), expand Method into the APA sub-sections — Participants, Materials, Procedure, Analysis — when the work warrants them.
4. Findings — the evidence, organized by theme or dimension. Each claim cites its source and, where evidence is contested, reports the verification status.
5. Discussion — interpretation, limitations, threats to validity, and open questions.
6. References — full citation list.

Citation style — selectable

Pick one mode and apply it consistently for the whole document:

• Author-date (APA, default) — in text, e.g. (Source, 2026), resolving to a full alphabetized reference list. Anchor APA to the 7th Edition (verify the current edition live).
• Numbered (Vancouver / IMRaD / ICMJE) — numbered in-text markers, e.g. [1], [2], resolving to a numerically ordered reference list. Use this for IMRaD/ICMJE-style work, which expects numbered (Vancouver) citations rather than author-date. Anchor to the current ICMJE Recommendations (updated roughly annually — verify the current revision live; do not bake a dated revision in).

Every reference derives from a genuinely verifiable source with a citation URL or DOI. No uncited claims, regardless of mode. Never fabricate a citation — omit a claim entirely rather than inventing a source for it.

Required figures & matter

• Front matter: title, author/attribution, date, abstract, optional table of contents.
• Figures: tables and figures as the evidence warrants — include a table whenever multiple findings are compared on shared attributes. Number and caption every figure; reference each in the text. Any figure, chart, or diagram is rendered as a fenced mermaid code block (never ASCII art, an image link, or Graphviz/DOT); a required figure is never silently omitted — if the data cannot support it, say so in prose. Plain tabular matter stays a Markdown table.
• Back matter: full References section; optional appendix for extended data or the method's verification log.

Rules

• Citation mode is selectable, not standard-certified. The IMRaD/ICMJE landscape anchor is a contested one — pick author-date (APA 7th) or numbered (Vancouver/ICMJE) and apply it consistently, but do not over-attribute strict standard conformance; verify the live edition of whichever you choose. APA Method sub-sections (Participants/Materials/ Procedure/Analysis) are an optional APA-mode expansion, not a default requirement.
• Every claim is traceable to a cited, real, verifiable source; no orphan facts and no fabricated citations.
• State limitations honestly — an undiscussed weakness is a defect, not an omission.
• Report the verification status of contested evidence; do not silently drop or launder disputed findings — annotate them (e.g. "preliminary evidence suggests", "contested"). Exclude claims that evidence has actively refuted.
• Hedge uncertain claims; present ranges when sources disagree.
• Exhaustive coverage: build the report from the full body of surviving evidence available — every finding worth including gets its own treatment (claim, citation, interpretation), never condensed to a cherry-picked subset. A silently dropped finding is a defect.
• MIF level: author at least MIF Level 1 (never bare, frontmatter-less prose); climb to Level 3 when the drafting context supplies provenance and citation detail.

Anti-triggers — do not use this genre for

• An informal, pre-build trade-off narrative — that is google-design-doc: conversational, alternatives reasoned in prose, no IMRaD structure or citation-style requirement.
• A practitioner decision report built around a mandatory options-vs-criteria comparison table — that is engineering: its center of gravity is the Trade-offs table, not scholarly method and cited evidence.
• A single already-made, immutable decision record — that is adr.

MIF frontmatter

type: semantic — an academic report is declarative knowledge (a scholarly claim set with method and evidence), not a time-bound event or a step sequence. Use entity_type: academic-paper and a namespace of the form academic/<area> (e.g. academic/information-retrieval). Climb to L2 with namespace, modified, title, and tags when the drafting context supplies them. Gate every output with mif-validate at its target level; the floor is --level 1.

Why machine-readable — the point of MIF here

Question an agent asks	Answered by (frontmatter)
Is this finding still current, or has it been superseded?	temporal.validFrom / ttl
Where did the evidence come from; can I trust it?	provenance (W3C-PROV) + citations[]
Which claims are settled vs. contested?	citations[] citationRole + hedged prose in Findings/Discussion
What prior work or companion artifact does this relate to?	typed relationships[] (e.g. relates-to a dataset or prior study)

The same document still reads as a human scholarly paper and projects losslessly to JSON-LD and back — one artifact, two readers.

The L1 -> L3 climb (two exemplars)

• templates/good-l1.md — L1 floor: id, type, created + body. A complete, valid academic report, but opaque to a machine consumer.
• templates/good.md — L3 (highest this genre supports): adds namespace, modified, temporal validity, W3C-PROV provenance, citations[] tied to the cited claims, and a typed relationships[] graph (e.g. relates-to a companion dataset or design doc). Validate with mif-validate --level 3.

Author at the highest level the drafting context supports (grade down rather than fabricate). templates/bad.md shows the antipattern: a report that states findings as fact with no citations to support them — this genre's single most load-bearing requirement.