Prompt and checklist templates that came out of our published methodology work. Each one links back to the article that explains why it is built the way it is.
These are convenience artifacts, not stand-alone resources. We extracted them from articles where the reasoning was already on the page, so a reader who wants to adopt one can grab the template and follow the link back when something does not quite fit. The rule we keep for ourselves: no template without an article, because the reasoning matters as much as the form.
Each card below links to the source article and to a plain Markdown file we can drop into a project folder.
A pre-registration scaffold for AI-assisted empirical work. Locks the primary specification before the data are touched, asks for a written justification on any deviation, and leaves a third-party-readable audit. Built on the rule, gate, verification pattern.
A weekly multi-source scan (SSRN, NBER, Semantic Scholar) consolidated into one pass, paired with a three-tier verification funnel for every citation that makes it into a manuscript. Designed to catch hallucinated papers, misattributed claims, and author swaps.
A five-minute capture routine for the end of a productive AI-assisted session. Four categories (what changed, why, what is next, what we tried that did not work), plus the one-sentence prompt we use to ask the agent to draft the summary for us.
A data card schema that tags each dataset by the questions it can answer, not just by what it contains. Includes two prompts, one for generating tags on intake and one for querying the inventory when a new research question surfaces.
Every template here has an accompanying article. That is deliberate. A prompt without its reasoning is easy to misapply, and the reasoning is what we want a reader to internalize. The template is the convenience layer, the article is the substantive one.
If we add a template later, it will follow the same rule: a published article first, then the extract. If something would be useful to us in our own work, let us know at [email protected].
New analysis delivered to your inbox. Policy-relevant findings with methodology. No spam, unsubscribe anytime.
Subscribe for Free