Module types

Source

Structs§

AgentFeedbackSummary: Response from GET /v1/agents/{id}/feedback/summary
AgentNetworkEdge: An edge between memories from two different agents
AgentNetworkInfo: Summary info for an agent in the cross-agent network
AgentNetworkNode: A memory node in the cross-agent network (includes agent_id)
AgentNetworkStats: Statistics for the cross-agent network
AgentStats: Agent memory statistics
AgentSummary: Lightweight agent summary for batch listing (uses count() not get_all)
AggregateFunctionInput: Wrapper for parsing aggregate function from JSON array
AggregationGroup: Single group in aggregation results
AggregationRequest: Request for aggregation query (Turbopuffer-inspired)
AggregationResponse: Response for aggregation query
BatchForgetRequest: Request for DELETE /v1/memories/forget/batch
BatchForgetResponse: Response from DELETE /v1/memories/forget/batch
BatchMemoryFilter: Filter predicates for batch memory operations.
BatchQueryItem: A single query within a batch request
BatchQueryRequest: Batch query request - execute multiple queries in parallel
BatchQueryResponse: Batch query response
BatchQueryResult: Results for a single query within a batch
BatchRecallRequest: Request for POST /v1/memories/recall/batch
BatchRecallResponse: Response from POST /v1/memories/recall/batch
BatchTextQueryRequest: Batch text query request
BatchTextQueryResponse: Response from batch text query
ColumnUpsertRequest: Column-based upsert request (Turbopuffer-inspired) All arrays must have equal length. Use null for missing values.
ConsolidateRequest: Request to consolidate related memories
ConsolidateResponse: Response from consolidation
CrossAgentNetworkRequest: Request for cross-agent memory network graph
CrossAgentNetworkResponse: Response from cross-agent network query
DecayConfig: Configuration for importance decay
DeduplicateRequest: Request to deduplicate memories
DeduplicateResponse: Response from deduplication
DefaultQuotaResponse: Response for default quota query
DeleteDocumentsRequest: Request to delete documents from full-text index
DeleteDocumentsResponse: Response from deleting documents
DeleteRequest: Delete request
DeleteResponse: Delete response
DuplicateGroup: A group of duplicate memories
EntityResult: A single extracted entity (shared with inference crate — mirrored here for API types).
ExportRequest: Request to export vectors from a namespace with pagination
ExportResponse: Response from export operation
ExportedVector: A single exported vector record
ExtractEntitiesRequest: Request to extract entities from content without storing. POST /v1/memories/extract
ExtractEntitiesResponse: Response from POST /v1/memories/extract and GET /v1/memories/{id}/entities
FeedbackHealthQuery: Query params for GET /v1/feedback/health
FeedbackHealthResponse: Response from GET /v1/feedback/health
FeedbackHistoryEntry: One recorded feedback event stored in memory metadata (feedback_history).
FeedbackHistoryResponse: Response from GET /v1/memories/{id}/feedback
FeedbackRequest: Request to provide feedback on a recalled memory (legacy — body contains memory_id)
FeedbackResponse: Response from feedback
ForgetRequest: Request to forget (delete) memories
ForgetResponse: Response from forget
FullGraphEdge: An edge in the full knowledge graph
FullGraphNode: A node in the full knowledge graph
FullKnowledgeGraphRequest: Request for full knowledge graph (all memories, pairwise similarity)
FullKnowledgeGraphResponse: Response from full knowledge graph query
FullTextIndexStats: Full-text index statistics
FullTextSearchRequest: Request to search for documents
FullTextSearchResponse: Full-text search response
FullTextSearchResult: Single full-text search result
GraphCluster: A cluster of related memories
GraphEdgeResponse: A single edge in a graph response.
GraphExportResponse: Response from agent graph export.
GraphNodeResponse: A single node in a graph traversal response.
GraphPathQuery: GET /v1/memories/:id/path
GraphPathResponse: Response from shortest-path query.
GraphStats: Statistics about the full knowledge graph
GraphTraverseQuery: GET /v1/memories/:id/graph
GraphTraverseResponse: Response from graph traversal.
HybridSearchRequest: Hybrid search request combining vector similarity and full-text search
HybridSearchResponse: Hybrid search response
HybridSearchResult: Single hybrid search result
IndexDocumentRequest: Request to index a document for full-text search
IndexDocumentsRequest: Request to index multiple documents
IndexDocumentsResponse: Response from indexing operation
KgExportJsonResponse: Response from GET /v1/knowledge/export (format=json)
KgExportParams: GET /v1/knowledge/export — export graph as JSON or GraphML
KgPathParams: GET /v1/knowledge/path — shortest path between two memory IDs
KgPathResponse: Response from GET /v1/knowledge/path
KgQueryParams: GET /v1/knowledge/query — JSON DSL for graph filtering/traversal
KgQueryResponse: Response from GET /v1/knowledge/query
KnowledgeGraphEdge: Knowledge graph edge
KnowledgeGraphNode: Knowledge graph node
KnowledgeGraphRequest: Request for knowledge graph traversal
KnowledgeGraphResponse: Response from knowledge graph query
ListSessionsResponse: Response listing sessions
Memory: A memory stored by an AI agent
MemoryFeedbackRequest: Request for POST /v1/memories/{id}/feedback (INT-1 — memory_id in path)
MemoryImportancePatchRequest: Request for PATCH /v1/memories/{id}/importance (INT-1 — memory_id in path)
MemoryLinkRequest: POST /v1/memories/:id/links — create an explicit edge
MemoryLinkResponse: Response from explicit link creation.
MemoryPolicy: Per-namespace memory lifecycle policy (COG-1).
MultiVectorSearchRequest: Request for multi-vector search with positive and negative vectors
MultiVectorSearchResponse: Response from multi-vector search
MultiVectorSearchResult: Single result from multi-vector search
NamespaceEntityConfigRequest: Request to update entity extraction config for a namespace. PATCH /v1/namespaces/{namespace}/config
NamespaceEntityConfigResponse: Response from PATCH /v1/namespaces/{namespace}/config
PaginationCursor: Internal cursor state for pagination
QueryRequest: Query request for vector search
QueryResponse: Query response
QuotaCheckRequest: Request to check if an operation would exceed quota
QuotaCheckResult: Quota check result
QuotaConfig: Quota configuration for a namespace
QuotaListResponse: Response listing all namespace quotas
QuotaStatus: Combined quota status showing config and current usage
QuotaUsage: Current quota usage for a namespace
RankByInput: Input format for rank_by that handles JSON array syntax Examples:
RecallRequest: Request to recall memories by semantic query
RecallResponse: Response from recall
RecallResult: Single recall result
SearchMemoriesRequest: Request for advanced memory search
SearchMemoriesResponse: Response from memory search
SearchResult: Single search result
Session: An agent session
SessionEndRequest: Request to end a session
SessionEndResponse: Response from ending a session
SessionMemoriesResponse: Response for session memories
SessionStartRequest: Request to start a session
SessionStartResponse: Response from starting a session
SetDefaultQuotaRequest: Request to set default quota configuration
SetQuotaRequest: Request to set quota for a namespace
SetQuotaResponse: Response from setting quota
StalenessConfig: Configuration for bounded staleness reads
StoreMemoryRequest: Request to store a memory
StoreMemoryResponse: Response from storing a memory
SummarizeRequest: Request to summarize memories
SummarizeResponse: Response from summarization
TextDocument: A text document with metadata for text-based upsert
TextQueryRequest: Request for text-based query (auto-embedded)
TextQueryResponse: Response from text-based query
TextSearchResult: Single result from text search
TextUpsertRequest: Request to upsert text documents (auto-embedded)
TextUpsertResponse: Response from text upsert operation
UnifiedQueryRequest: Unified query request with rank_by parameter (Turbopuffer-inspired)
UnifiedQueryResponse: Unified query response with $dist scoring
UnifiedSearchResult: Single result from unified query
UpdateImportanceRequest: Request to update importance of a memory
UpdateMemoryRequest: Request to update a memory
UpsertRequest: Request to upsert vectors
UpsertResponse: Response from upsert operation
Vector: A vector with associated metadata
WakeUpResponse: Response from GET /v1/agents/{agent_id}/wake-up (DAK-1690).

Enums§

AggregateFunction: Aggregate function for computing values across documents
DecayStrategy: Strategy for importance decay
DistanceMetric: Distance metric for vector comparison
EmbeddingModelType: Available embedding models.
FeedbackSignal: Feedback signal for active learning
FilterCondition: Comparison operators for filter conditions
FilterExpression: A filter expression that can be a single field condition or a logical combinator
FilterValue: A filter value that can be compared against metadata fields
MemorySortField: Fields to sort memories by
MemoryType: Type of memory stored by an agent
QuotaEnforcement: Quota enforcement mode
RankBy: Ranking function for unified query API Supports vector search (ANN/kNN), full-text BM25, and attribute ordering
ReadConsistency: Read consistency level for queries (Turbopuffer-inspired)
SortDirection: Sort direction for attribute ordering
VectorSearchMethod: Vector search method

Type Aliases§

NamespaceId: Namespace identifier
VectorId: Unique identifier for a vector

Module types

Module types Copy item path

Structs§

Enums§

Type Aliases§

Module types