review-product-value

Review Product Value Skill

[!IMPORTANT] Review realized product value against registered measurement contracts and baseline evidence, separating delivery completion from causal ROI.

Optional args: slug=, ticket=<id/url>, mode=interactive|autonomous|channel, channel=, auto_continue=true|false, profile=business|hybrid|technical.

Instructions

When the user asks to perform this workflow, execute the following steps:

Review Product Value Workflow

Goal: Evaluate post-release product value by comparing observed outcomes against registered metrics and investment assumptions without inferring causal ROI from delivery alone.

Steps

1. Preflight and intake:
   • Preflight: verify product-management/product-evidence, product-management/product-metrics, product-management/product-investment; conditional product-management/product-roadmap only if proposing ordering changes. Scripts: scripts/validate-product-evidence.mjs (ordinary <root>/product-management/product-evidence, Kiro <root>/product-management-product-evidence), scripts/evaluate-investment.mjs (ordinary <root>/product-management/product-investment, Kiro <root>/product-management-product-investment). Validator invocation: node <validator> <ledger.json> --authority-root <canonical-root> --as-of <UTC> --gate release (or structural). Calculator invocation: node <calculator> <scenario.json>. Exit: 0 = coherent (structural, readiness not_assessed) or gate eligible for host check only; never authorization; 3 = proposal_only/gap; 1/2/4 = error (never validated). Emit exact gap strings (<pkg> not installed; opt in at a reviewed ref and sync, product-management/product-evidence excluded; opt in at a reviewed ref and sync, validate-product-evidence.mjs missing; repair package, evaluate-investment.mjs missing; repair package, Node >=20 unavailable; calculation not performed, analytics evidence unavailable; value inconclusive) with safe next action; preserve delivery feature_status. See product-evidence Verification Workflow.
   • Intake: slug, scope, original investment ref (historical attribution only, not action authority), registered MET contracts, and comparison window. No automated background scheduler.
2. Inspect measurement contracts and evidence:
   • Audit MET contracts (AC-009): verify status (planned|instrumented|observed|unavailable; no immature status enum), exact query_revision = query origin revision, eligibility/cohort, baseline XOR scoped baseline_exception (purpose: "release", outcome: "risk_accept"), measurement_owner, and review_trigger. On failure, record precise missing_evidence gap and return inconclusive, never invented observations.
   • Retrieve observed SRC data and actual incremental costs. Check displaced work / opportunity cost rationale. Retain paired financial input/result (financial_input, financial_result / horizon) separately.
3. Assess attribution and guardrails:
   • Inspect guardrails (churn, errors, latency). If guardrail harm or regression is detected, escalate immediately to authorized incident controls (incident-hotfix); agents cannot autonomously roll out or revert.
   • Evaluate causal attribution: classify as causal_supported | descriptive | unknown. Document confounders (marketing, seasonality, pricing) and counterfactuals. scale and causal_supported require observed metrics plus sources. Never infer causal ROI from release.
4. Formulate advisory recommendation:
   • Emit REV record with recommendation: scale | iterate | stop | inconclusive. Record explicit limitations. REV.decision is historical attribution only; any capital, roadmap, or retirement action requires a separate recorded value_review / host decision.
   • Recommendations are advisory; emit host decision needed for any capital, roadmap, or retirement action.
   • Preserved delivery invariant: keep delivery feature_status (e.g. released or verified) unchanged even when product value is inconclusive; never set blocked for product-evidence gaps alone.