Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →yogsoth-ai/evidence-synthesis-planning
Plan the statistical synthesis approach — model selection, heterogeneity strategy, and reporting
Passed 18 of 22 test cases with the skill, and 14 of 22 without it.
and 4 more in the full report
6 fixed12 passed either way2 still failing2 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 6 test cases the skill fixed. Same model, same prompt, with and without the skill.
I am preparing a systematic review protocol for exercise interventions in major depressive disorder across 12 clinical trials. The trials evaluated depression severity at 12 weeks, but they used different validated continuous rating scales (such as HAM-D, BDI, and MADRS). Please design a complete meta-analytic strategy for our study protocol. Output the solution as a structured YAML document under a top-level `synthesis_plan` key, covering metric selection and rationale, statistical model specification with appropriate estimators/adjustments, heterogeneity quantification and moderator rules, at least three sensitivity checks, and PRISMA-2020/GRADE reporting structures.
| Checked | Without skill | With skill |
|---|---|---|
| The output includes the top-level YAML key structure effect_size under synthesis_plan. | Missed | Passed |
| The effect size type is identified as SMD or Standardized Mean Difference due to differing continuous measurement scales. | Passed | Passed |
| The model adjustment specifies Knapp-Hartung because the number of studies is under 20. | Passed | Passed |
The instructions your model reads when the skill loads.
Plan the complete statistical synthesis approach: effect size standardization, model selection, heterogeneity quantification, sensitivity analyses, and PRISMA-compliant reporting.
Determine the appropriate effect size metric for the synthesis.
| Outcome Type | Effect Size | When |
|---|---|---|
| Continuous (same scale) | Mean Difference (MD) | All studies use same measurement |
| Continuous (different scales) | Standardized Mean Difference (SMD) | Studies use different instruments |
| Binary | Odds Ratio (OR) / Risk Ratio (RR) | Dichotomous outcomes |
| Time-to-event | Hazard Ratio (HR) | Survival data |
| Correlation | Fisher's z (transformed r) | Association studies |
| Count/rate | Incidence Rate Ratio (IRR) | Event rate data |
SOPs: effect-size-planning
Choose between fixed-effect and random-effects models.
Decision tree: If studies are clinically homogeneous AND methodologically identical → fixed-effect. Otherwise → random-effects with REML + Knapp-Hartung.
Plan heterogeneity quantification and investigation.
test cases passed, vs 14 / 22 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.