Use when planning or auditing the analysis of an American Journal of Sociology (AJS) manuscript so the evidence credibly supports the theoretical claim. Covers analysis norms, uncertainty, robustness, and triangulation across quantitative, comparative-historical, and ethnographic work. Improves the analysis chain; it does not fabricate results.
SkillScore82Maintained
Automatic
◆Research pack
Last verified 8/15/2026
What it changes
Passed 18 of 22 test cases with the skill, and 13 of 22 without it.
Without skill59%
With skill82%
Checks it passed that the model alone missed
The output specifies `ajs-tables-figures` under the 【Next】 field.
The 【Next】 field specifies `ajs-tables-figures` as the recommended next step.
The output specifies calling `audit_result(result_id)` to list missing checks and suggested functions.
and 7 more in the full report
8 fixed10 passed either way1 still failing3 broke
Pass-rate lift
+23 pp
Tokens
+43%
Agent turns
0%
Response time
-21%
Model gemini-3.6-flashMeasured 8/15/2026v118 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 8 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 8
Prompt
Before providing final, submission-ready advice on an empirical manuscript for AJS, what external resource must be checked to confirm live data policies and volatile journal rules?
What was checked, without and with the skill
Checked
Without skill
With skill
The output requires checking `resources/official-source-map.md` before delivering submission-ready advice.
Missed
Passed
case-21 · 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.md105 linessingle file
Data Analysis (ajs-data-analysis)
At AJS the analysis exists to make the theoretical claim credible — not to display technique. A
generalist, double-blind reviewer will ask whether the evidence actually warrants the claim and whether
candor about uncertainty is present. This skill stress-tests the analysis chain in the idiom of your
work.
When to trigger
Planning the analysis, or auditing it before writing up
A reader doubts robustness, the evidence-to-claim link, or the handling of uncertainty
Reconciling multiple methods or data sources into one coherent argument
Deciding which analyses are confirmatory vs. exploratory
Analysis norms (by tradition)
Quantitative
Report uncertainty honestly (intervals, not just stars); avoid implying causality the design
cannot support.
Show that results are not artifacts: principled robustness (alternative specifications,
samples, measures), not a fishing expedition; keep seeds and pinned versions.
Distinguish preregistered/confirmatory from exploratory analyses where applicable.
Comparative-historical
Make the inferential logic explicit (necessary/sufficient conditions, sequence, conjuncture);
show how disconfirming evidence was sought and weighed.
Cite primary sources so a reader could follow the trail.
Ethnographic / interview
Show the analytic procedure: how codes/themes were built, how negative cases were handled, how
representativeness within the case is judged.
Quote enough to let the reader assess the inference from data to claim.
Triangulation (an AJS strength)
AJS often rewards convergent evidence — a mechanism shown through more than one window
(e.g., statistics + cases, or interviews + administrative data). When methods disagree, say so and
theorize the discrepancy rather than hiding it.
Referee-pushback patterns on the evidence chain (AJS fixes)
Use when executing and reporting the analysis for an American Journal of Political Science (AJPS) manuscript. AJPS will have a third-party verifier re-run your exact code against the numerical results in the main text before publication, so analyze reproducibly from the first line. Guides analysis and reporting norms; it does not fabricate results.