Use when building tables and figures for an American Educational Research Journal (AERJ) manuscript. Exhibits must be self-contained, accessible, and formatted to APA 7th edition, and they must be anonymized for masked review. Improves exhibits; it does not generate data.
SkillScore95Maintained
Automatic
◆Research pack
Last verified 8/15/2026
What it changes
Passed 21 of 22 test cases with the skill, and 16 of 22 without it.
Without skill73%
With skill95%
Checks it passed that the model alone missed
The response specifies adding operational definitions and exemplar quotations for each theme.
The response specifies including information on bandwidth or pre-trend testing in the plot or notes.
The response requires removing the district identifier ('Chicago Public Schools') to maintain masked review.
and 2 more in the full report
5 fixed16 passed either way1 still failing0 broke
Pass-rate lift
+23 pp
Tokens
+52%
Agent turns
0%
Response time
-23%
Model gemini-3.6-flashMeasured 8/15/2026v121 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 5 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 5
Prompt
I am preparing a regression table for a multilevel study on student math performance across 45 school districts for submission to AERJ. The table displays Level-1 and Level-2 predictors, random slope variances, and school-level ICCs. Please audit this table draft and provide your review using the required output format:
【Exhibit】 table / figure and a short purpose statement
【Self-contained?】 Y/N assessment of whether title and notes explain sample size, units, abbreviations, and significance
【APA 7th format】 Y/N indicator
【Uncertainty shown】 Y/N check for standard errors, confidence intervals, and explicit estimands
【Accessible】 Y/N evaluation of colorblind safety and grayscale legibility
【Masked】 Y/N check ensuring no institutional or site identifiers remain
【Next】 aerj-writing-style
What was checked, without and with the skill
Checked
Without skill
With skill
The response presents the review in the exact bracketed text block structure without converting field tags into Markdown headings.
Missed
Passed
The response includes all seven required fields from 【Exhibit】 through 【Next】.
Passed
Passed
case-01 · gemini-3.6-flash · 8/15/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md102 linessingle file
Tables & Figures (aerj-tables-figures)
In AERJ, a reader should grasp each exhibit without hunting through the text, and every exhibit must
follow APA 7th-edition table/figure conventions. Because review is masked, exhibits must carry
no identifying clues (institution names, data-source giveaways in titles).
When to trigger
Designing tables/figures or revising them after analysis
A reviewer found an exhibit unreadable, mislabeled, or non-APA
Preparing exhibits for a masked submission (no identifying content)
Deciding what belongs in the body vs an appendix
Principles
Self-contained. Title states what the exhibit shows; notes define every abbreviation, the
sample/N, the units, and significance/probability conventions. A reader should not need the text.
APA 7th-edition format. Table number + italic title; horizontal rules only (no vertical rules);
figure number + title; a general note, then specific notes, then probability notes.
Report estimates with uncertainty. Coefficients with SEs/CIs; for multilevel models, label
levels and random effects; mark the estimand, not just stars.
Figures earn their place. Use a figure when it shows something a table cannot (trajectories,
interactions, distributions, RD/event-study, joint displays for mixed methods). Otherwise use a table.
Accessibility. Colorblind-safe palettes; legible in grayscale; readable at print size; vector
output for line art.
Qualitative exhibits. Coding trees, thematic maps, and evidence tables (theme → exemplar
quotation) are legitimate exhibits — anonymize participants and sites.
Match the package. Exhibit numbers/values must match the analysis output and the deposited
materials (see aerj-transparency-and-data-policy).
Execution bridge (StatsPAI / Stata MCP)
Generate exhibits from the fitted result, not by retyping numbers (the usual source of
body-vs-supplement drift). Full map: execution-with-mcp. AERJ is empirical education research — field experiments and observational school data; multilevel inference and many-outcome corrections are central.
Tables:etable (multi-model columns) or did_summary_to_latex straight from the
result_id.
Figures:plot_from_result / enhanced_event_study_plot / event_study_table —
axis units and the SE/clustering note baked in.
Every note names the estimator + clustering and states the effect size in
Use when building tables and figures for an American Journal of Political Science (AJPS) manuscript. AJPS counts table/figure headers and notes toward the word limit, and the deposited code must regenerate every exhibit exactly for third-party verification. Designs self-contained, reproducible exhibits; it does not run the analysis.