Types Reference

All types are importable from @hebbs/sdk. They are implemented as TypeScript interfaces with readonly fields and enums with string values.

Memory

The core data type representing a stored memory.

Field	Type	Description
`id`	`Buffer`	16-byte unique memory identifier
`content`	`string`	The memory text
`importance`	`number`	Importance score (0.0—1.0)
`context`	`Record<string, unknown>`	Structured key-value metadata (used by analogical recall for structural matching)
`entityId`	`string \| undefined`	Entity scope. `undefined` if stored without an entity.
`createdAt`	`number`	Creation timestamp in microseconds
`updatedAt`	`number`	Last modification timestamp in microseconds
`lastAccessedAt`	`number`	Last access timestamp in microseconds
`accessCount`	`number`	How many times this memory has been recalled (Hebbian reinforcement)
`decayScore`	`number`	Current decay score (importance adjusted by time decay)
`kind`	`MemoryKind`	Memory classification (episode, insight, revision)
`embedding`	`number[]`	The embedding vector (empty unless explicitly requested)

MemoryKind

Enum classifying the type of memory.

Variant	Value	Description
`EPISODE`	`'episode'`	A stored event or observation — the default kind
`INSIGHT`	`'insight'`	Generated by the reflection pipeline from clusters of episodes
`REVISION`	`'revision'`	Created by `revise()` — supersedes a previous memory
`UNSPECIFIED`	`'unspecified'`	Unknown or not set

Edge

A directed relationship between two memories. Used with remember() to create graph edges.

Field	Type	Description
`targetId`	`Buffer`	16-byte ID of the target memory
`edgeType`	`EdgeType`	Relationship classification
`confidence`	`number \| undefined`	Strength of the relationship (0.0—1.0). Default: `undefined` (engine default).

import { EdgeType } from '@hebbs/sdk';

const edge = {
  targetId: mem.id,
  edgeType: EdgeType.FOLLOWED_BY,
  confidence: 0.95,
};

EdgeType

Enum for relationship types between memories. Used in Edge and for filtering causal recall via RecallStrategyConfig.edgeTypes.

Variant	Value	Description
`CAUSED_BY`	`'caused_by'`	Target caused this memory
`RELATED_TO`	`'related_to'`	General semantic relationship
`FOLLOWED_BY`	`'followed_by'`	This memory followed the target chronologically
`REVISED_FROM`	`'revised_from'`	Revision lineage (set automatically by `revise()`)
`INSIGHT_FROM`	`'insight_from'`	Insight derived from source memories (set by `reflect()`)
`UNSPECIFIED`	`'unspecified'`	Unknown or not set

RecallStrategy

Enum for the four recall strategies.

Variant	Value	Description
`SIMILARITY`	`'similarity'`	Semantic vector similarity (HNSW index)
`TEMPORAL`	`'temporal'`	Time-proximity ordering (B-tree index)
`CAUSAL`	`'causal'`	Graph traversal following causal edges
`ANALOGICAL`	`'analogical'`	Cross-domain pattern matching (blends embedding + structural similarity)

RecallStrategyConfig

Per-strategy configuration for advanced recall tuning. See Client Reference for the full parameter table and usage examples.

Field	Type	Default	Description
`strategy`	`string`	(required)	Strategy name
`entityId`	`string \| undefined`	`undefined`	Override entity scope for this strategy
`topK`	`number \| undefined`	`undefined`	Per-strategy result limit
`efSearch`	`number \| undefined`	`50`	HNSW candidate count (similarity)
`timeRange`	`[number, number] \| undefined`	`undefined`	`[startUs, endUs]` bounds (temporal)
`seedMemoryId`	`Buffer \| undefined`	`undefined`	Graph traversal start node (causal)
`edgeTypes`	`EdgeType[] \| undefined`	`undefined`	Restrict traversal edges (causal)
`maxDepth`	`number \| undefined`	`5`	Max hops (causal, hard cap: 10)
`analogicalAlpha`	`number \| undefined`	`0.5`	Structural vs. embedding blend (analogical)

ScoringWeights

Override composite scoring weights for recall() and prime().

Field	Type	Default	Description
`wRelevance`	`number \| undefined`	`0.5`	Weight for semantic relevance (cosine similarity)
`wRecency`	`number \| undefined`	`0.2`	Weight for temporal recency
`wImportance`	`number \| undefined`	`0.2`	Weight for stored importance value
`wReinforcement`	`number \| undefined`	`0.1`	Weight for recall frequency (Hebbian reinforcement)
`maxAgeUs`	`number \| undefined`	`undefined`	Max age for recency decay (microseconds). Server default: 30 days.
`reinforcementCap`	`number \| undefined`	`undefined`	Cap for reinforcement scaling. Server default: 100.

const weights: ScoringWeights = {
  wRelevance: 0.8,
  wRecency: 0.1,
  wImportance: 0.05,
  wReinforcement: 0.05,
};

StrategyDetail

Per-strategy metadata for a recall result. Each strategy populates different fields.

Field	Type	Description
`strategy`	`string`	Strategy name: `'similarity'`, `'temporal'`, `'causal'`, `'analogical'`
`relevance`	`number`	Raw relevance score for this strategy
`distance`	`number \| undefined`	HNSW distance (similarity only)
`timestamp`	`number \| undefined`	Memory timestamp in microseconds (temporal only)
`rank`	`number \| undefined`	Ordering rank in temporal results (temporal only)
`depth`	`number \| undefined`	Graph traversal depth from seed (causal only)
`embeddingSimilarity`	`number \| undefined`	Embedding similarity component (analogical only)
`structuralSimilarity`	`number \| undefined`	Structural similarity component (analogical only)

RecallResult

A single memory returned from a recall query, with composite score and strategy details.

Field	Type	Description
`memory`	`Memory`	The recalled memory
`score`	`number`	Composite score (weighted blend of relevance, recency, importance, reinforcement)
`strategyDetails`	`StrategyDetail[]`	Per-strategy metadata. A memory found by multiple strategies has multiple entries.

StrategyError

Reports a per-strategy error when one strategy fails but others succeed.

Field	Type	Description
`strategy`	`string`	Strategy that failed
`message`	`string`	Error message

RecallOutput

The complete response from a recall() call.

Field	Type	Description
`results`	`RecallResult[]`	Ranked list of recalled memories (sorted by composite score)
`strategyErrors`	`StrategyError[]`	Errors from individual strategies, if any (partial success is possible)

PrimeOutput

Response from a prime() call.

Field	Type	Description
`results`	`RecallResult[]`	Blended temporal + similarity results
`temporalCount`	`number`	Number of results from the temporal component
`similarityCount`	`number`	Number of results from the similarity component

ForgetResult

Response from a forget() call.

Field	Type	Description
`forgottenCount`	`number`	Number of memories erased
`cascadeCount`	`number`	Number of related records (edges, snapshots) cascade-deleted
`tombstoneCount`	`number`	Number of tombstone records created for audit
`truncated`	`boolean`	Whether the result set was truncated (large entity deletions)

ReflectResult

Response from a reflect() call.

Field	Type	Description
`insightsCreated`	`number`	Number of new insight memories generated
`clustersFound`	`number`	Number of memory clusters identified
`clustersProcessed`	`number`	Number of clusters processed by the LLM pipeline
`memoriesProcessed`	`number`	Total number of source memories analyzed

SubscribePush

A push notification received on a subscription stream.

Field	Type	Description
`subscriptionId`	`number`	ID of the subscription that generated this push
`memory`	`Memory`	The surfaced memory
`confidence`	`number`	Confidence score for this push (0.0—1.0)
`pushTimestampUs`	`number`	Server timestamp when the push was generated (microseconds)
`sequenceNumber`	`number`	Monotonically increasing sequence for ordering

HealthStatus

Server health information returned by health().

Field	Type	Description
`serving`	`boolean`	Whether the server is serving requests
`version`	`string`	Server version string (e.g. `'0.1.0'`)
`memoryCount`	`number`	Total memories stored across all entities
`uptimeSeconds`	`number`	Server uptime in seconds