Skip to main content

Module diagnostics

Module diagnostics 

Source
Expand description

Stack Visualization, Diagnostics, and Reporting

ML-driven system for visualizing, diagnosing, and reporting on the health of the Sovereign AI Stack. Implements Toyota Way principles for observability.

§Toyota Way Principles

  • Mieruka (Visual Control): Rich ASCII dashboards make health visible
  • Jidoka: ML anomaly detection surfaces issues automatically
  • Genchi Genbutsu: Evidence-based diagnosis from actual dependency data
  • Andon: Red/Yellow/Green status with stop-the-line alerts
  • Yokoten: Cross-component insight sharing via knowledge graph

Re-exports§

pub use super::diagnostics_ml::ErrorForecaster;
pub use super::diagnostics_ml::ForecastMetrics;
pub use super::diagnostics_ml::IsolationForest;

Structs§

Anomaly
Detected anomaly in the stack
ComponentMetrics
Quality and performance metrics for a component
ComponentNode
A component in the stack knowledge graph
GraphMetrics
Computed graph-level metrics
HealthSummary
Summary of stack health for dashboard
StackDiagnostics
Main diagnostics engine for stack analysis

Enums§

AndonStatus
Andon board status for the entire stack
AnomalyCategory
Categories of anomalies
HealthStatus
Health status for components (Andon-style visual control)

Functions§

render_dashboard
Render diagnostics as ASCII dashboard