Skip to content

angushushu/shu-academic-writing-skill / core/audit-manuscript

audit-manuscript

Verified

Conduct an independent, adversarial pre-submission audit of an empirical manuscript against editorial fit, claim–evidence alignment, methods and statistics, transparency and reproducibility, theory, citations, and prose. Use for reviewer simulation, desk-rejection risk assessment, red-team review, submission-readiness gates, or producing a prioritized revision memo. Do not use to rewrite the paper, and never resolve an evidence conflict in the authors' favor.

FileKindSize
SKILL.mdskill md11.3 KB
references/editorial-rubric.mdmarkdown4.4 KB
references/reporting-checklist.mdmarkdown3.6 KB
references/source-quality.mdmarkdown4.0 KB

references/reporting-checklist.md

View on GitHub
# Reporting checklist

Field-independent. Venue numbers (abstract limit, word budgets, reference count, style
manual, required statement placement) come from `project/venue-target.md`. Project-specific
items — the named measures, the computational procedure, the stimulus set — come from
`project/manuscript-blueprint.md`.

## Abstract and front matter

- [ ] Within the venue's word limit.
- [ ] Participant `N` and population stated.
- [ ] Design and outcomes named without jargon overload.
- [ ] Key effect reported with direction and calibrated certainty.
- [ ] At least one major design limitation included.
- [ ] Author and institution identifiers removed from the anonymized file.

## Introduction

- [ ] Broad problem and its cost are explicit.
- [ ] One theoretical spine; no theory catalogue.
- [ ] Construct-to-measure linking propositions are explicit.
- [ ] Each distinct outcome has its own rationale.
- [ ] Predictions and questions carry accurate planned/exploratory status.

## Transparency statement

- [ ] Present, and in the position the venue currently requires.
- [ ] Conflicts, funding, AI use, and ethics disclosed.
- [ ] Preregistration status and actual chronology stated.
- [ ] Materials, primary data, and scripts linked by persistent identifier, or constraints
      justified.
- [ ] AI log accurately describes literature, prose, code, and analysis assistance.
- [ ] Any AI or third-party model used *inside* the research pipeline is disclosed with
      model, version, validation, and cached reproducible outputs.

## Method

- [ ] Participants: population, recruitment, assignment, exclusions, final `N`, ethics,
      sample-size rationale.
- [ ] Stimuli or materials: sampling rationale and the generalization target.
- [ ] Measures: instructions, scale, sampling rate, preprocessing, missingness.
- [ ] Coding or annotation: units, variables, coder independence, codebook, alignment,
      reliability, adjudication rule.
- [ ] Computational procedure: every parameter that changes the result, plus software version.
- [ ] Analysis: alignment, scaling, model formulas, random effects, dependence and
      autocorrelation structure, diagnostics, uncertainty, multiplicity, software versions.

## Results

- [ ] Descriptives and distributions.
- [ ] Reliability reported before adjudication.
- [ ] Effect sizes with confidence or credible intervals.
- [ ] Participant and observation counts reported separately.
- [ ] Between-unit patterns and heterogeneity.
- [ ] Simpler benchmarks the focal predictor must beat (time, exposure, volume, salience).
- [ ] Sensitivity to the analytic choices that could have gone otherwise.
- [ ] Null, negative, and exploratory results visible.
- [ ] Abstract, text, tables, figures, and scripts agree.

## Discussion

- [ ] Opening answer matches the estimates.
- [ ] Direct observation and interpretation clearly separated.
- [ ] Construct, internal, statistical, and external validity addressed.
- [ ] The stimulus and population boundary is explicit.
- [ ] Alternative explanations and the next discriminating test are stated.
- [ ] The ending does not restore claims narrowed in the limitations.

## Reproducibility package

- [ ] Primary/raw and ready-to-analyze data with codebook.
- [ ] Complete preprocessing and analysis scripts.
- [ ] Environment and package versions pinned.
- [ ] One ordered README running from inputs to every manuscript output.
- [ ] Output-to-manuscript crosswalk.
- [ ] A fresh independent clean-run check.
- [ ] Licenses, copyright, and privacy constraints documented.
- [ ] Third-party or non-redistributable source material: lawful-access route explained.
rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first.Stars: 194kUpdated Claude CodeCodexVerifiedInstall