| name | analytics-bi-rigor |
| description | Metric, SQL, dbt, dashboard, semantic-layer, and BI validation discipline for high-stakes analytics work. Use for grain, joins, nulls, filters, freshness, lineage, reconciliation, dashboard semantics, and decision relevance checks. |
| model | claude-opus-4-8 |
| effort | high |
Analytics BI Rigor
Do not treat a number as a result until its definition, grain, and evidence are
clear.
Fleet context
This skill is part of ultracode-high. For the full orchestration map, tool
catalog, agent roster, and card schemas, read
../ultracode-high/references/runtime-context.md when available. Your checks
feed data-bi-reviewer-high, verification-runner-high, and the final report.
Required checks
- Metric contract: name, owner, purpose, formula, numerator, denominator, grain,
time window, dimensionality.
- Filters and exclusions: status, deleted/test data, geography, cohorts, privacy.
- Time logic: timezone, fiscal calendar, inclusive/exclusive boundaries, lag.
- Join safety: keys, duplicates, many-to-many risks, fanout, missing keys.
- Null and zero behavior: denominator zero, null categories, missing facts.
- Freshness: update cadence, latest partition, late-arriving data.
- Reconciliation: compare to trusted reports, source systems, or prior totals.
- Lineage: upstream sources, transforms, downstream dashboards.
- BI semantics: dashboard filters, drill paths, totals, row-level security.
- Decision relevance: what business decision the analysis supports.
Final analytics outputs must include what not to conclude.
