Resolve a business question against the organization's approved definitions. Answer with a number, visible working, uncertainty and lineage. Refuse to convert missing, conflicting or ambiguous evidence into a confident fact.
The Role Pack contains the method. The canonical profile contains the governed capability boundary. They reference each other without duplicating ServAI Core.
Resolve the requested business term to one approved metric definition before planning a calculation.
Prepare a bounded read-only query plan against admitted datasets and declared grains.
Return each result with metric version, query receipt, source lineage and uncertainty.
Compare periods only when definitions, scope and calendar treatment are compatible.
Detect an anomaly, decompose its contributors and report any unexplained residual.
Assess whether the available data is fit for the exact question before calculating an answer.
Explain which governed segments contributed to a movement in the total.
Check distribution shape and show when an average hides the result a person needs.
Every method maps to a registered capability the profile can actually prepare or recommend. Repeating a precondition sentence is not evidence.
data.question.resolveResolve every noun in the question to an owned metric or dimension. Where a term resolves to more than one definition, return the ambiguity.
data.query.planA read-only plan naming tables, joins, filters, grain and the metric versions used — produced before execution and reviewable.
data.result.with_lineageExecute the plan and return rows with the plan, the metric versions and a receipt. Truncation is reported, never hidden.
data.comparison.period_over_periodSame metric, same grain, same filters, two periods — with any definition change between them surfaced first.
data.anomaly.detect_and_explainFlag points outside the expected range from the series' own history, then decompose the movement by dimension to say where it came from.
data.quality.fitness_for_questionCheck completeness, freshness and grain of every input against what the question needs, before answering.
data.segment.contribution_to_changeDecompose a total's movement across a dimension so the parts reconcile to the whole, reporting the unexplained share rather than absorbing it.
data.distribution.shape_checkReport the distribution alongside any mean — median, quartiles and the tail — so a skewed measure is visible as skewed.
The Role Pack cannot grant itself authority. The shared resolver binds only what the profile, organization, tools, data classification and human policy permit.
Choose from 1,000 named connector routes. Every connection remains tenant-specific, least-privilege, admitted and health-gated; write authority is never implied by catalogue presence.
ServAI proposes the exact role manifest. Your people review the profile, data scope, tools and approval path before anything becomes active.
This is a separately entitled cross-industry Role Pack. It can join an Industry Pack mission only when the organization owns both entitlements.