Enterprise Credibility Audit

Capability Readiness Center

This page audits bp Sphere capabilities the way BP technology, finance, audit, security, operations, and transformation leaders will judge them: business value, runtime explainability, governance, operability, and North Star movement.

Data boundary: scores below are computed at request time from the live FastAPI route map, BP RBAC config, source modules, and available runtime/API modules. Missing production telemetry is shown as a gap; it is not filled with invented values.
Five Enterprise Credibility Audits
AuditQuestionRequired Proof
Business AuditDoes it solve a real finance problem?Problem, KPI, beneficiary, decision speed, risk reduction, cost reduction.
Runtime AuditCan we explain how it works?Signal, context, policy, evidence, agent, decision, action, learning.
Governance AuditCan we prove it is controlled?Policies, human approvals, agent permissions, security gates, audit trail.
Operations AuditCan we operate it at scale?Latency, failures, queue depth, fallback path, cost controls, ownership.
Transformation AuditDoes it move BP toward the North Star?Current state, future state, role change, KPI impact, North Star alignment.
Live Runtime Signals Used For This Audit
SignalObserved ValueSourceStatus
Capability readiness manifest10tenant domains/finance/platform/capability_manifests.yamlAvailable
Release gate minimum score80capability_readiness.release_gateAvailable
FastAPI route map8Live request.app.routesAvailable
Evidence Vault APITrueiris_sor/api/evidence_vault.pyAvailable
Decision Replay APITrueiris_sor/api/decision_replay.pyAvailable
SOR Adapter APITrueiris_sor/api/sor_adapters.pyAvailable
Mission Runtime Standard ServiceTrueiris_sor/services/mission_runtime_standard_service.pyAvailable
Agent Manifest ServiceTrueiris_sor/services/manifest_service.pyAvailable
Tenant RBAC route policyTrueconfig/tenants/bp/rbac/iris_rbac_menu.yamlAvailable
Capability Readiness Scorecard

Scores are not manually entered. They are derived from route registration, RBAC registration, module existence, source-level proof terms, and platform module availability. A low score means the code/runtime proof is absent or not yet wired.

CapabilityBusinessAgentDataEvidenceSecurityCostOperationsTransformationAudit NoteEvidence Basis
Future Finance Transformation Journey671001001007510075100Computed from live checks. Missing: route, production telemetry: current_state_metrics,value_realization_actuals.route=missing; RBAC=yes; module=yes; runtime=agent_manifest_service,cost_ui,decision_replay_api,evidence_api
Continuous Close Intelligence671001001007510075100Computed from live checks. Missing: route, production telemetry: sap_journal_feed,blackline_reconciliation_feed,close_calendar_feed.route=missing; RBAC=yes; module=yes; runtime=agent_manifest_service,cost_ui,decision_replay_api,evidence_api
Enterprise Runtime Topology67100100100751007575Computed from live checks. Missing: route, production telemetry: runtime_health,adapter_health,model_usage.route=missing; RBAC=yes; module=yes; runtime=agent_manifest_service,cost_ui,decision_replay_api,evidence_api
Technology Credibility Audit671001001007510075100Computed from live checks. Missing: route, production telemetry: runtime_health,adapter_health,security_events.route=missing; RBAC=yes; module=yes; runtime=agent_manifest_service,cost_ui,decision_replay_api,evidence_api
Agent Registry & Governance67100100100751007575Computed from live checks. Missing: route, production telemetry: agent_execution_history,certification_state.route=missing; RBAC=yes; module=yes; runtime=agent_manifest_service,cost_ui,decision_replay_api,evidence_api
Evidence Fabric / Evidence Vault67100100100751007575Computed from live checks. Missing: route, production telemetry: evidence_pack_count,artifact_count,integrity_status.route=missing; RBAC=yes; module=yes; runtime=agent_manifest_service,cost_ui,decision_replay_api,evidence_api
Context Graph67100100100751007575Computed from live checks. Missing: route, production telemetry: graph_object_count,relationship_count,source_freshness.route=missing; RBAC=yes; module=yes; runtime=agent_manifest_service,cost_ui,decision_replay_api,evidence_api
Cost Governance Center67100100100751007575Computed from live checks. Missing: route, production telemetry: token_usage,model_cost,sap_reads.route=missing; RBAC=yes; module=yes; runtime=agent_manifest_service,cost_ui,decision_replay_api,evidence_api
Transformation Heatmap671001001007510075100Computed from live checks. Missing: route, production telemetry: baseline_metrics,target_metrics,realized_value.route=missing; RBAC=yes; module=yes; runtime=agent_manifest_service,cost_ui,decision_replay_api,evidence_api
Local ServiceNow Runtime (Enterprise Work)673310010025677550Computed from live checks. Missing: route, RBAC, cost telemetry/content, production telemetry: sn_work_item_volume,approval_cycle_time,sla_breach_rate.route=missing; RBAC=missing; module=yes; runtime=agent_manifest_service,cost_ui,decision_replay_api,evidence_api
North Star Audit
CapabilityNorth Star DimensionCredibility Proof
Continuous CloseTimeClose readiness, journal readiness, reconciliation readiness, critical path, forecasted close duration.
Evidence FabricAssuranceEvidence packs, policies, transactions, documents, approvals, replay.
Context GraphDataUnified context across supplier, customer, contract, journal, asset, policy, decision, action.
Forecast AgentInsightDrivers, assumptions, scenarios, confidence, executive narrative, recommended action.
Policy AgentAssuranceAuthority limits, SOX, approval matrix, write-back boundary, escalation path.
Decision CenterWorkJudgment over processing through decision crates, supervisor approvals, and action queues.
Agent Credibility Audit
DimensionWhat Must Be VisibleWhere It Shows
IdentityAgent ID, version, owner, criticality, autonomy level.Agent Registry drawer.
ActivationWhy the agent ran and which event or signal triggered it.Runtime trace / signal feed.
InputsSource systems, APIs, tables, documents, context objects.Data lineage drawer.
PoliciesGovernance rules and action boundaries applied.Policy result tab.
EvidenceRecords, documents, calculations, approvals, and historical patterns used.Evidence pack viewer.
OutputsRecommendation, confidence, alternatives, impact, action payload.Decision crate.
EscalationSupervisor path, approval owner, override reason.Human accountability panel.
LearningOutcome captured, override captured, pattern updated.Learning and replay panel.
Runtime Audit Flow

Every credible capability should map to this flow. This prevents the demo from becoming a collection of widgets.

SignalWhat changed and why it matters.
ContextWhich enterprise objects, owners, history, and relationships apply.
PolicyWhat is allowed, blocked, or requires escalation.
EvidenceWhat proves the recommendation is justified.
AgentWhich agents participated and what each contributed.
DecisionRecommendation, confidence, alternatives, impact, owner.
ActionDraft, approval, write-back, workflow, communication, or escalation.
LearningOutcome, override, lesson, pattern update, replay.
Data Lineage Audit
DimensionRequirementUI Proof
SourceECC, CFIN, S4, Ariba, BlackLine, Databricks, treasury, documents, market feeds.Shown per object and per evidence record.
Access MethodRepresentative workshop adapter now; production target API, event, table, or curated Databricks product.Adapter detail drawer.
RefreshReal-time event, near-real-time sync, hourly, daily, or static policy/document version.Freshness badge.
ConfidenceSource quality, lineage completeness, evidence coverage, policy coverage.Confidence and data-quality score.
LineageSource -> context -> decision -> action.Lineage timeline.
Security, Cost, and Operations Audit
AuditQuestionRequired UI Proof
SecurityWho accessed it? Which agent executed? What permissions were used? What systems and sensitive data were touched?Identity drawer, agent identity, permission scope, security gateway result.
CostHow expensive is this capability to operate?Context reads, SAP reads, verification calls, writebacks, model usage, token cost, cache hit rate.
OperationsWhat happens if a dependency fails?Fallback path for agent failure, SAP unavailable, Databricks unavailable, model unavailable, policy denial.
Transformation Audit
DimensionQuestionExample Proof
Current StateHow work is done today.Manual investigation, process queues, periodic reporting, delayed controls.
Future StateHow work changes.Continuous decisions, preventive controls, evidence on demand, human supervision.
AgentsWho performs investigation and coordination.Named agent family, owner, inputs, outputs, policies.
Human RoleWhat changes for people.Processor -> decision analyst; coordinator -> decision orchestrator; controller -> risk supervisor.
KPI ImpactWhat improves.Cycle time, close duration, DSO, touchless rate, control coverage, forecast accuracy.
North Star ImpactWhich dimension moves.Time, work, posture, assurance, data, insight.
Immediate Audit List Before BP Demo
Page / CapabilityAudit FocusStatusGap To CloseEvidence Basis
Future FinanceBusiness problem, North Star alignment, KPI impact, transformation dashboard.PASSComputed from live checks. Missing: route, production telemetry: current_state_metrics,value_realization_actuals.route=missing; RBAC=yes; module=yes; runtime=agent_manifest_service,cost_ui,decision_replay_api,evidence_api
Continuous CloseReadiness score logic, evidence, escalations, replay.PASSComputed from live checks. Missing: route, production telemetry: sap_journal_feed,blackline_reconciliation_feed,close_calendar_feed.route=missing; RBAC=yes; module=yes; runtime=agent_manifest_service,cost_ui,decision_replay_api,evidence_api
Context GraphObject lineage and data sources.PARTIALComputed from live checks. Missing: route, production telemetry: graph_object_count,relationship_count,source_freshness.route=missing; RBAC=yes; module=yes; runtime=agent_manifest_service,cost_ui,decision_replay_api,evidence_api
Agent RegistryIdentity, permissions, policies, owners, runtime metrics.PARTIALComputed from live checks. Missing: route, production telemetry: agent_execution_history,certification_state.route=missing; RBAC=yes; module=yes; runtime=agent_manifest_service,cost_ui,decision_replay_api,evidence_api
Runtime TopologySecurity, recovery, cost controls, adapters, replay.PARTIALComputed from live checks. Missing: route, production telemetry: runtime_health,adapter_health,model_usage.route=missing; RBAC=yes; module=yes; runtime=agent_manifest_service,cost_ui,decision_replay_api,evidence_api
Cost GovernanceSAP usage, Databricks usage, LLM usage, deterministic work split.PARTIALComputed from live checks. Missing: route, production telemetry: token_usage,model_cost,sap_reads.route=missing; RBAC=yes; module=yes; runtime=agent_manifest_service,cost_ui,decision_replay_api,evidence_api
Transformation HeatmapCurrent state, future state, gap, required capabilities.PASSComputed from live checks. Missing: route, production telemetry: baseline_metrics,target_metrics,realized_value.route=missing; RBAC=yes; module=yes; runtime=agent_manifest_service,cost_ui,decision_replay_api,evidence_api