Provenance Trail

• Complete data lineage: Track every step from data ingestion through AI processing to final execution
• Cryptographic verification: Each step includes hash verification for immutable audit trails
• Actor tracking: Distinguish between system, AI, and human actions with visual indicators
• Model versioning: Track which AI models and versions were used at each step
• Data quality metrics: Monitor data quality scores and row counts throughout the pipeline
• Searchable timeline: Filter and search through audit trail events
• Export capabilities: Export provenance data in JSON, PDF, or CSV formats

Import the ProvenanceTrail component and provide an array of steps representing your audit trail. Each step should include actor, action, timestamp, and optional data sources, outputs, and verification hashes.

import { ProvenanceTrail } from '@/components/ai/provenance-trail'

const steps = [
  {
    id: 'step-1',
    timestamp: '2024-12-01T10:30:00Z',
    action: 'data_ingestion',
    actor: { type: 'system', name: 'Data Pipeline' },
    target: 'Analysis Job',
    sources: [
      {
        name: 'sales-data.csv',
        rows: 45000,
        quality_score: 0.98,
        hash: 'sha256:abc123...'
      }
    ],
    hash: 'sha256:verified...'
  },
  // ... more steps
]

<ProvenanceTrail
  steps={steps}
  visualizationMode="timeline"
  searchable={true}
  verifiable={true}
  onExport={(format) => exportData(format)}
  onVerify={(stepId, verified) => handleVerification(stepId, verified)}
/>