govern-tables-figures — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited govern-tables-figures (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
In a Governance paper the exhibits are where a comparative reviewer checks whether the institutional claim actually travels. A single-country result dressed up as a wall of coefficients reads as parochial; a clean cross-national table or a coefficient plot that shows where the effect holds and where it does not reads as a contribution to how the state governs. With a working cap of ≤ 9,000 words excluding citations/bibliography (检索于 2026-06；以官网为准), every exhibit must earn its place, and the heavy robustness grids belong in the supplementary file, not the article.
alone — units, sample, time window, country/case set, N, and what the estimate is. State the estimand, not just "Model 3."
countries, regimes, policy domains, or institutional types. Order rows/panels by the institutional logic (e.g., by state capacity, by Westminster vs. consensus), not alphabetically, so the pattern reads off the page.
predicted-probability plots, and event-study plots communicate magnitude and uncertainty better than a coefficient dump. Always show intervals, never stars alone.
rules. SEs in parentheses under coefficients; report N, the FE/clustering structure, and the estimator; define every abbreviation in the note.
series by shape/linetype as well as color so the print and the PDF both work.
replication materials cited in the Data Availability Statement exactly (see govern-transparency-and-data).
group by regime type or region; flag missing data honestly rather than blanking cells.
ideal for showing where a mechanism holds and where scope conditions bite.
the left of zero, effect periods on the right, reference period marked.
use a sequential colorblind-safe ramp, classed not continuous, with the breaks justified in the note.
argument.
diagrams, and evidence tables linking each claim to its source (pairs with QDR-style documentation).
Generate exhibits from the fitted result, not by retyping numbers (the usual source of body-vs-supplement drift). Full map: execution-with-mcp. Governance is public administration and institutions research — comparative and causal designs on governance reforms; the chain serves its quantitative-causal lane, while comparative-historical / qualitative work uses its own standards.
etable (multi-model columns) or did_summary_to_latex straight from theresult_id.
plot_from_result / enhanced_event_study_plot / event_study_table —axis units and the SE/clustering note baked in.
interpretable units.
See a full fitted-result → exhibit chain in the JF execution walkthrough.
【Main exhibit】what it shows + why this form (table/plot/map)
【Comparison visible?】cross-country/institutional contrast reads off the page? [Y/N]
【Self-contained?】caption + labels + note + units/N/case-set present? [Y/N]
【Accessible?】grayscale-legible + colorblind-safe? [Y/N]
【Article vs supplementary】split decided; 9,000-word impact noted
【Reproducible?】generated by master script, matches materials? [Y/N]
【Next】govern-writing-style../../resources/external_tools.md — plotting/mapping packages and comparative-governance data sources../../resources/official-source-map.md — word cap, figure/format guidelines, supplementary-material policy~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.