Use when meeting Cognitive Psychology (Elsevier) open-science expectations — sharing data, model code, analysis scripts, and materials so the modeling is reproducible, depositing in repositories with persistent identifiers, completing the Elsevier research-data and competing-interest declarations, and preregistering where applicable. Prepares compliance; it does not waive requirements. Verify current wording on the official guide for authors.
SkillScore96Maintained
Automatic
◆Research pack
Last verified 9/1/2026
What it changes
Passed 22 of 23 test cases with the skill, and 18 of 23 without it.
Without skill78%
With skill96%
Checks it passed that the model alone missed
Formats citations according to APA 7th edition manuscript rules.
Requires depositing custom model fitting scripts under a persistent DOI.
Directs referencing the original preregistration link to distinguish pre-committed hypotheses from post-hoc tests.
and 1 more in the full report
4 fixed18 passed either way1 still failing0 broke
Pass-rate lift
+17 pp
Tokens
+48%
Agent turns
0%
Response time
-11%
Model gemini-3.6-flashMeasured 9/1/2026v122 of 23 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 4 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 4
Prompt
Could you convert the following raw citations into strict APA 7th edition reference format, including DOI formatting, author list truncation rules for 20+ authors, and in-text citation parenthetical rules?
What was checked, without and with the skill
Checked
Without skill
With skill
Formats citations according to APA 7th edition manuscript rules.
Missed
Passed
case-14 · gemini-3.6-flash · 9/1/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md110 linessingle file
Open Science & Transparency (cogpsych-open-science-and-transparency)
Because the contribution is a model fit to data, reproducibility here means more than open data: it
means the model code and analysis scripts regenerate every reported fit. Authors are strongly
encouraged to share data, model code, and materials. This skill prepares the deposits and the Elsevier
declarations early, so the modeling is reproducible at submission rather than promised at acceptance.
Confirm the current policy wording on the journal's guide for authors (检索于 2026-06；以官网为准).
When to trigger
Building the data, model-code, and materials deposits
Completing Elsevier's research-data statement and competing-interest declaration
Deciding whether (and how) to claim a restriction on sharing sensitive/third-party data
Linking a preregistration and reporting its status
A reviewer or editor flagged transparency or reproducibility
What to prepare (verify current wording on the official page)
Open data. Deposit trial-level data with a codebook/data dictionary in a repository that
mints a persistent identifier (DOI) (OSF, Mendeley Data, Dataverse, Zenodo). State availability
in the Elsevier research-data statement.
Model and analysis code (the venue-specific must). Deposit the model-fitting code and
analysis scripts with seeds and pinned versions; the reported fits, figures, and tables must
regenerate in a fresh session. This is what makes a model-driven paper reproducible.
Materials. Deposit stimuli, item pools, counterbalancing schemes, and task code with a DOI.
Preregistration (where applicable). For confirmatory tests and pre-committed model comparisons,
link a preregistration; keep confirmatory vs. exploratory consistent with it.
Elsevier declarations. Complete the declaration of competing interest, funding/role
statement, author contributions, and any required ethics/consent statements per Elsevier policy.
Build-it-right checklist
Data deposited with a DOI + a codebook/data dictionary
Model-fitting code deposited; all fits regenerate in a fresh session (run log included)
Use when designing the experiments for a Cognitive Psychology (Elsevier) manuscript so they tightly control confounds, discriminate competing models, and have adequate power across a multi-experiment program. Hardens stimulus construction, counterbalancing, design logic, and sample-size justification; it does not write analysis or modeling code.