
This is the skill you reach for after your research question is approved but before you touch any outcome data. It forces you to write down the entire analysis plan as if you're handing it to someone with zero context: which datasets, how each variable is computed, the exact model or test, which confounds you're adjusting for and how, the decision rules, power justification, all broken into two to five minute steps with validation checks. The plan gets saved to a dated markdown file and includes a mandatory simulated data validation where your pipeline has to recover a known effect before you trust it on real data. It's opinionated about avoiding researcher degrees of freedom and makes you pre-register before execution, which is the point.
$ npx -y skills add k-dense-ai/science-superpowers --skill designing-the-analysis --agent claude-codeInstalls into .claude/skills of the current project.
Write a comprehensive analysis plan assuming the analyst has zero context for this project and questionable statistical taste. Document everything they need: which datasets and variables, how each construct is computed, the exact model or test, the sample size / power justification, which confounds are handled and how, the decision rules, and the planned figures. Give them the whole thing as bite-sized steps. DRY. YAGNI. Pre-register. Validate pipelines on known data. Commit frequently.
Assume they are a capable programmer but know almost nothing about this domain, this dataset, or good statistical design.
Announce at start: "I'm using the designing-the-analysis skill to create the analysis plan."
Save plans to: docs/science-superpowers/plans/YYYY-MM-DD-<topic>.md
If the question document still bundles several independent investigations, stop and split it — one plan per question. Each plan should produce an interpretable, self-contained result.
Sponsors