Analyze datasets for evidence-backed decisions
data-analysisskillsetup L1★14
akillness/oh-my-skills ↗Causal-lift measurements
data-analysis+8pp vs no-skill baselinewith-skill 88% · baseline 80%
Measured by running the task with and without this artifact, K=5, graded by deterministic checks — no LLM judging.
What it does
Analyze datasets to support decisions with evidence-backed conclusions
Best for
When you have structured business data and need decision-quality analysis that separates observation from interpretation.
Inputs
- · Dataset or export (CSV, JSON, SQL table)
- · Decision question to answer
- · Performance metrics and KPIs to analyze
Outputs
- · Analysis narrative with findings and caveats
- · Evidence-based conclusions and next actions
Preconditions
- · Data has been cleaned and quality-checked
- · Decision question is clearly framed
Failure modes
- · Jumps to conclusions without data quality triage
- · Conflates correlation with causation
- · Uses wrong analysis lane for the dataset size
Trust signals
- · 5-step reasoning framework (frame → profile → lane → separate → evidence)
- · Data-quality triage checklist
- · Lane selection matrix for spreadsheet/SQL/notebook/summary
- · Pattern recognition examples