---
name: research-documentation
description: >-
  Use when a repository produces measurements, benchmarks or experiments that
  need writing up as research rather than as a guide, or when any document in
  it has to be authored in LaTeX and compiled to PDF. What counts as research
  and what does not, the docs/research layout, the shape of a study document,
  the evidence rules (every number traceable, reproduction stated as a
  command, validate the instrument before trusting it, negative results get
  written up), when a study graduates to a LaTeX paper or a Beamer deck, why
  the house preamble is a file every document inputs rather than a block each
  one copies, how a non-study document borrows the house style, how Mermaid
  diagrams reach a PDF from one source and are called by slug, how language
  and variant editions come from one trunk, what a LaTeX .gitignore covers,
  the workflow that builds any of them, and the check that keeps it from
  drifting. Bundled: the study and paper templates, the house preamble, the
  Beamer theme and the check.
---

# Scientific research documentation

**Read [`references/00-RESEARCH-DOCUMENTATION.md`](references/00-RESEARCH-DOCUMENTATION.md) before applying any of this.**
That file is the standard; everything below it is a summary to help you decide
whether this skill applies and to check your work afterwards.

Reference-architecture principles: P14.

Bundled templates: [`assets/TEMPLATE.md`](assets/TEMPLATE.md), [`assets/HOUSE-PREAMBLE.tex`](assets/HOUSE-PREAMBLE.tex), [`assets/PAPER-TEMPLATE.tex`](assets/PAPER-TEMPLATE.tex), [`assets/BEAMER-THEME.sty`](assets/BEAMER-THEME.sty), [`assets/CHECK-PAPERS.mjs`](assets/CHECK-PAPERS.mjs).

## What this standard covers

- What counts as research (and what doesn't)
- Where research lives in a repository
- The shape of a study
- The evidence rules
- From study to paper (LaTeX)
- The house preamble is a file, not a block
- Presenting work as slides (Beamer)
- Documents that borrow the house style without being studies
- Diagrams in a PDF
- Publishing a document in more than one language
- One trunk, several editions
- Building the PDF in CI
- What a `.gitignore` for LaTeX actually has to cover
- Checking the papers
- Relationship to the rest of the standards
- Worked example

## What counts as research (and what doesn't)

A **study** asks a question whose answer wasn't known before the work was
done, and answers it with evidence: measurements, verified computations,
reproducible experiments, or systematic analysis of committed code. Examples:
benchmarking an algorithm against its theoretical complexity, verifying an
implementation against independently derived expected values, characterizing
a system's behavior under a sweep of inputs, root-causing a discrepancy
between a model's specification and its implementation.

Not research, and living elsewhere:

- **Theory/background material** — teaching a reader established results
  (textbook algorithms, probability rules). That is a theory doc
  (`docs/THEORY.md` in `bayesian-inference` is the worked example) and a
  study *cites* it rather than repeating it.
- **Design decisions** — those go in the decisions log
  (`docs/architecture/05-DECISIONS.md` pattern).
- **Gap analysis / roadmap** — `docs/architecture/` per the constitution.

---

Generated from [`docs/research/00-RESEARCH-DOCUMENTATION.md`](https://github.com/konradcinkusz/architecture-standards/blob/main/docs/research/00-RESEARCH-DOCUMENTATION.md) by `scripts/build-marketplace.mjs`. Do not edit this file: change the source document, or its entry in `catalog/marketplace.catalog.json`, and re-run the generator.