Targeted Blast Radius Audit — IRONTRUST Financials (GATEKEEPER PROTOCOL)
Scope: Scale Selector + Live Financial Aggregation on Risk Ops Dashboard.
Mode: Read-only. No code changes.
1. Financial Data Flow — File Map
1.1 Prisma schema (risk/loss as BigInt)
| File | Role | Current state |
|---|
prisma/schema.prisma | Defines Company.industry_avg_loss_cents, Company.infrastructure_val_cents (BigInt?); ActiveRisk.score_cents (BigInt); ThreatEvent.financialRisk_cents (BigInt); Tenant.ale_baseline (BigInt). | Ready — All financial fields are BigInt (cents). |
1.2 API routes (dashboard data)
| File | Role | Current state |
|---|
app/api/dashboard/route.ts | GET /api/dashboard: tenant-scoped companies, active_risks (score_cents), threatEvents. Serializes companies and risks to JSON. | Needs refactor — Serialization converts BigInt → Number: score_cents: Number(risk.score_cents), industry_avg_loss_cents: ... Number(...), infrastructure_val_cents: ... Number(...). Safe for typical GRC scale (e.g. < $10B) but TAS violation for values beyond Number.MAX_SAFE_INTEGER; prefer string or BigInt-safe serialization for future scale. |
1.3 Zustand store (financial totals & scale)
| File | Role | Current state |
|---|
app/store/riskStore.ts | Holds acceptedThreatImpacts ($M per threat), dashboardLiabilities, riskOffset, currencyMagnitude / currencyScale (AUTO | K |
1.4 UI — Left sidebar (Current Risk / GRC Gap) & Scale Selector
| File | Role | Current state |
|---|
app/components/StrategicIntel.tsx | Left sidebar: “YOUR CURRENT RISK”, “POTENTIAL IMPACT”, “GRC GAP”. Uses acceptedThreatImpacts, pipelineThreats, currencyMagnitude, formatRiskExposure. Two parallel aggregates: (1) exactTotalCurrentRisk from acceptedThreatImpacts; (2) totalRiskMillions from pipelineThreats.reduce(..., loss ?? score). Bar labels use totalRiskMillions; trend chart uses dynamicCurrentRisk (from accepted + pipeline). GRC Gap = getIndustryMetrics(selectedIndustry).grc — hardcoded strings ($4.3M, $5.1M, etc.). | Needs refactor — (1) Dead/duplicate logic: Two different “current risk” definitions (accepted sum vs pipeline sum); bar display uses pipeline-only sum, so “Current Risk” label may not match accepted-only semantics. (2) Mocked static: getIndustryMetrics() returns fixed $8.5M / $4.3M etc.; GRC Gap is not live. |
app/components/TopNav.tsx | Scale Selector (AUTO, K, M, B, T): reads currencyScale, calls setCurrencyScale(scale). | Ready — Wired to riskStore; no financial math. |
app/components/ExecutiveSummary.tsx | Uses currencyMagnitude, formatRiskExposure(adjustedLiabilityExposureUsd, currencyMagnitude) for baseline/current/mitigated. Aggregates from acceptedThreatImpacts, pipelineThreats, totalMitigatedRiskM. | Ready — Uses scale and formatting correctly. |
app/components/LiabilityExposureDisplay.tsx | Single liability display: formatRiskExposure(totalUsd, currencyMagnitude). | Ready — Thin wrapper over store + riskFormatting. |
app/components/DashboardAlertBanners.tsx | Regulatory alerts; shows formatRiskExposure(alert.liabilityM * 1_000_000, currencyMagnitude) per alert. | Ready — Scale-aware display. |
app/components/ExecutiveSummary.tsx (line 130) | Label “Current Risk” in UI. | Ready — Display only. |
Note: No separate Left_Sidebar.tsx; left-side content is in StrategicIntel.tsx.
1.5 Irontrust math utilities
| File | Role | Current state |
|---|
core/irontrust/ale-engine.ts | calculateRiskLevel(aleCents: bigint, thresholdCents: bigint) → CRITICAL | ELEVATED |
app/utils/riskFormatting.ts | formatRiskExposure(value: number, scale: CurrencyMagnitude) → string (e.g. "5.0M"). Input value in dollars (not cents). AUTO picks K/M/B/T by magnitude. | Ready — Input is number (dollars); scale selector drives display. No BigInt in this file (accepts already-converted number). |
2. Vulnerabilities & Dead Code (within above files only)
| Location | Issue | Severity |
|---|
app/api/dashboard/route.ts (lines 78, 86–87) | BigInt → Number serialization for score_cents, industry_avg_loss_cents, infrastructure_val_cents. Precision loss for values > 2^53 − 1. | TAS violation (use BigInt-safe serialization for currency). |
app/components/StrategicIntel.tsx (lines 579–587) | getIndustryMetrics(selectedIndustry) returns hardcoded $8.5M, $4.3M, etc. GRC Gap is not driven by live data. | Mocked static — Refactor when wiring live financial aggregation. |
app/components/StrategicIntel.tsx (lines 527–529 vs 591–594) | Two “current risk” aggregates: exactTotalCurrentRisk (accepted) and totalRiskMillions (pipeline). Bar labels use totalRiskMillions; chart uses dynamicCurrentRisk. Risk of confusion or inconsistency. | Dead/duplicate logic — Clarify single source of truth for “Current Risk” (accepted vs pipeline vs both). |
app/store/riskStore.ts | PipelineThreat.loss and score are number ($M). No unused variables in the listed slice; store is consistent. | — |
app/utils/riskFormatting.ts | No unused variables; no currency in Float/Number violation (input is already dollar number). | — |
core/irontrust/ale-engine.ts | No dead code; BigInt-only. | — |
3. Test Coverage Mapping
| Component / utility | Test file(s) | What’s covered |
|---|
| GET /api/dashboard (tenant isolation, payload shape) | tests/integration/dashboard.test.ts | 401 without tenant; 401 empty x-tenant-id; 200 with Vaultbank context, companies/risks shape, no cross-tenant bleed. Prisma mock includes $transaction. |
| Dashboard E2E (shell, layout) | tests/e2e/dashboard.spec.ts | Dashboard main visible, GRC gate / drawer behavior. |
| Dashboard load / pipeline | tests/e2e/stage1-validation.spec.ts, tests/e2e/stage1-validation-simplified.spec.ts, tests/e2e/threatPipeline.spec.ts | Dashboard-main wait, pipeline/registration flows (some tests skipped). |
| ALE engine (BigInt risk levels) | core/irontrust/ale-engine.test.ts | Baselines (Medshield, Vaultbank, Gridcore), boundaries, financial guardrails (negative/zero), monotonic invariant. |
| riskFormatting.ts | — | No dedicated unit tests. |
| riskStore (currency scale, financial aggregates) | — | No dedicated unit tests. Only indirect coverage via E2E. |
| StrategicIntel / ExecutiveSummary (scale selector + formatRiskExposure) | — | No unit tests; only E2E that hit dashboard. |
4. Terminal commands to test this logic before changes
Run these before implementing Scale Selector + Financial Aggregation:
# Dashboard API (tenant isolation, payload shape, serialization)
npx vitest run tests/integration/dashboard.test.ts
# ALE engine (BigInt risk level math)
npx vitest run core/irontrust/ale-engine.test.ts
# E2E: dashboard shell + pipeline (chromium only)
npx playwright test tests/e2e/dashboard.spec.ts --project=chromium
# Optional: full E2E suite for dashboard/pipeline (includes skipped tests)
npx playwright test tests/e2e/dashboard.spec.ts tests/e2e/threatPipeline.spec.ts tests/e2e/stage1-validation.spec.ts tests/e2e/stage1-validation-simplified.spec.ts --project=chromium
Recommended before wiring live aggregation: Add unit tests for:
app/utils/riskFormatting.ts — formatRiskExposure(value, scale) for AUTO/K/M/B/T and edge cases.app/store/riskStore.ts — setCurrencyMagnitude / setCurrencyScale and any financial aggregate used by the new feature.
5. TAS COMPLIANCE CHECK
| Directive | Status |
|---|
| Trace financial data flow | Done — Prisma (BigInt), dashboard API, riskStore, StrategicIntel/TopNav/ExecutiveSummary/LiabilityExposureDisplay/DashboardAlertBanners, ale-engine, riskFormatting. |
| Prisma schema (risk/loss BigInt) | Done — Documented; all relevant fields are BigInt. |
| API routes (dashboard) | Done — app/api/dashboard/route.ts; BigInt→Number serialization flagged. |
| Zustand store (financial totals) | Done — app/store/riskStore.ts; scale + aggregates documented. |
| UI: Left sidebar (Current Risk / GRC Gap) | Done — StrategicIntel.tsx; duplicate logic and hardcoded GRC Gap flagged. |
| UI: Scale Selector (AUTO/K/M/B/T) | Done — TopNav.tsx; wired to riskStore. |
| Irontrust math (ale-engine) | Done — core/irontrust/ale-engine.ts; BigInt-only. |
| Identify vulnerabilities & dead code | Done — Within listed files: TAS (BigInt→Number), mocked GRC Gap, two “current risk” definitions. |
| Test coverage mapping | Done — dashboard.test.ts, ale-engine.test.ts, E2E specs; riskFormatting and riskStore untested. |
| Exact terminal commands | Done — Listed in §4. |
Result: Audit complete. Dashboard API serialization and StrategicIntel GRC Gap/current-risk logic are the main refactor targets before wiring Scale Selector and live Financial Aggregation.