Interface: IngestionStatsDto

@kortexya/reasoninglayer / IngestionStatsDto

Interface: IngestionStatsDto

Defined in: src/types/ingestion.ts:869

Ingestion statistics.

Remarks

Comprehensive statistics from the ingestion pipeline including chunk processing, entity extraction, deduplication, relations, and quality metrics.

Properties

alignmentStats?

optional alignmentStats?: AlignmentStatsDto

Defined in: src/types/ingestion.ts:879

Source-text alignment statistics.

alreadyIngested

alreadyIngested: boolean

Defined in: src/types/ingestion.ts:877

true when these bytes were already ingested in the tenant: nothing was extracted again, and IngestionStatsDto.documentTermId names the earlier registry document that holds them (the first document with these bytes). The request’s collectionPath / previousDocumentId were applied to that document; an intent the engine cannot apply is refused, never dropped.

chunksFailed

chunksFailed: number

Defined in: src/types/ingestion.ts:881

Number of chunks that failed extraction.

chunksProcessed

chunksProcessed: number

Defined in: src/types/ingestion.ts:883

Number of text chunks processed.

clausesExtracted?

optional clausesExtracted?: number

Defined in: src/types/ingestion.ts:885

Number of contract clauses extracted (obligations, permissions, …).

communityStats?

optional communityStats?: CommunityStatsDto

Defined in: src/types/ingestion.ts:887

Community detection statistics (GraphRAG-style clustering).

documentTermId?

optional documentTermId?: string | null

Defined in: src/types/ingestion.ts:894

Document-registry id (UUID) of the document that holds this content: the document term the linking stage created, or, when IngestionStatsDto.alreadyIngested, the earlier document with the same bytes. Absent when linking was skipped.

entitiesCreated

entitiesCreated: number

Defined in: src/types/ingestion.ts:896

Number of new entities created.

entitiesExtracted

entitiesExtracted: number

Defined in: src/types/ingestion.ts:898

Number of entities extracted from text.

entitiesFilteredLowQuality?

optional entitiesFilteredLowQuality?: number

Defined in: src/types/ingestion.ts:900

Number of entities filtered out due to low-quality names.

entitiesMerged

entitiesMerged: number

Defined in: src/types/ingestion.ts:902

Number of entities merged (deduplicated).

entitiesPendingReview

entitiesPendingReview: number

Defined in: src/types/ingestion.ts:904

Number of entities pending human review.

entitiesRefused

entitiesRefused: RefusedEntityDto[]

Defined in: src/types/ingestion.ts:909

The extracted entities the declaration verdict refused, so they were not written. Every refusal is listed; empty when none was refused.

entityMergeAudit?

optional entityMergeAudit?: EntityMergeAuditDto[]

Defined in: src/types/ingestion.ts:911

The entity merges the pipeline made, each with its evidence.

estimatedCostUsd?

optional estimatedCostUsd?: number | null

Defined in: src/types/ingestion.ts:913

Estimated USD cost of LLM API calls (if model pricing is available).

failedChunks?

optional failedChunks?: ChunkFailureDto[]

Defined in: src/types/ingestion.ts:915

Details of failed chunks (first 5).

groundingStats?

optional groundingStats?: GroundingStatsDto

Defined in: src/types/ingestion.ts:917

Entity grounding statistics (linking to external ontologies).

model?

optional model?: string | null

Defined in: src/types/ingestion.ts:919

LLM model used for extraction.

ontologyVerifier?

optional ontologyVerifier?: OntologyVerifierStatsDto

Defined in: src/types/ingestion.ts:921

Statistics of the learned ontology verifier.

pipelineQuality?

optional pipelineQuality?: PipelineQualityStatsDto

Defined in: src/types/ingestion.ts:923

Pipeline quality statistics (inference rules, validation, feedback loop).

processingTimeMs

processingTimeMs: number

Defined in: src/types/ingestion.ts:925

Processing time in milliseconds.

qaPairsGenerated?

optional qaPairsGenerated?: number

Defined in: src/types/ingestion.ts:927

Number of Q&A pairs that document enrichment generated.

referencesResolved

referencesResolved: number

Defined in: src/types/ingestion.ts:929

Number of references resolved.

referencesUnresolved

referencesUnresolved: number

Defined in: src/types/ingestion.ts:931

Number of unresolved references (residuated).

relationsExtracted

relationsExtracted: number

Defined in: src/types/ingestion.ts:933

Number of relations extracted from text.

relationsIntegrated

relationsIntegrated: number

Defined in: src/types/ingestion.ts:935

Number of relations integrated as features on entities.

relationsOrphaned?

optional relationsOrphaned?: number

Defined in: src/types/ingestion.ts:937

Number of suspended relations that stayed unresolved.

relationsResumed?

optional relationsResumed?: number

Defined in: src/types/ingestion.ts:939

Number of suspended relations resumed when their endpoints appeared.

relationsSuspended?

optional relationsSuspended?: number

Defined in: src/types/ingestion.ts:941

Number of relations suspended because an endpoint was not yet available.

sortNames

sortNames: string[]

Defined in: src/types/ingestion.ts:943

Names of created/reused sorts.

sortsCreated

sortsCreated: number

Defined in: src/types/ingestion.ts:945

Number of new sorts created.

sortsReused

sortsReused: number

Defined in: src/types/ingestion.ts:947

Number of existing sorts reused.

sotaExtractionGate?

optional sotaExtractionGate?: SotaExtractionGateStatsDto

Defined in: src/types/ingestion.ts:949

Statistics of the extraction quality gate.

termIds

termIds: string[]

Defined in: src/types/ingestion.ts:951

IDs of created/updated terms.

tokenUsage?

optional tokenUsage?: TokenUsageDto

Defined in: src/types/ingestion.ts:953

Token usage from LLM API calls during this ingestion.