observational-analysis-rigor

Plugin observational-analysis-rigor in wan-huiyan/overnight-workflows. Commands read from the repo's manifest — 1 month ago.

Skills (31)

• amplifying-an-existing-number-is-a-provenance-recheck-trigger
• blind-rederive-pass-when-orchestrator-already-read-the-answer
• cohort-broadening-event-source-scope-leak
• cohort-milestone-lift-is-funnel-position-not-effect
• conditional-funnel-by-segment-within-level
• coverage-limited-join-validate-unbiased-before-trusting
• data-provenance-verifier
• differentiator-scoping-by-provenance-not-signal
• finding-verification-live-bq-triple-probe
• frozen-cohort-rebucket-newer-model-contemporaneity-leak
• funnel-lever-vs-predictor-deleaked-forward-gap
• funnel-stage-lift-needs-downstream-capacity-check
• metric-window-spec-needs-event-time-distribution-probe
• ml-feature-pit-derive-from-anchor-else-ablation-upper-bound
• near-absorbing-state-rate-deflated-by-snapshot-residue
• null-bucket-hides-progressors-in-snapshot-training
• numeric-rederive-confirms-value-not-label-or-cohort
• observational-analysis-rigor
• observational-version-comparison-confounded-by-time
• pit-history-reconstruction-needs-canonical-code-source-of-truth
• pit-panel-first-appearance-left-truncation-incident-anchor
• ship-the-correction-to-every-rendered-surface
• single-anchor-point-estimate-needs-range-framing
• sister-cohort-high-overlap-inflates-aggregate-sum
• snapshot-churns-out-one-outcome-class-biases-the-rate
• snapshot-feature-pit-leak-rate-measure-history-change-after-target
• snapshot-transient-state-cohort-censors-outcome-use-ever-entered-from-history
• target-leakage-guard-misses-transforms-of-the-target
• two-lever-synergy-scale-dependent-and-simpson-ceiling
• verifier-rederive-from-raw-not-the-checked-artifact
• within-stratum-residual-event-floor-anchor-split