Skip to main content

Crate yantrikdb

Crate yantrikdb 

Source

Re-exports§

pub use engine::YantrikDB;
pub use engine::tenant::TenantManager;
pub use engine::tenant::TenantConfig;
pub use engine::graph_state::CognitiveGraphSaveResult;
pub use engine::graph_state::CognitiveGraphStats;
pub use engine::graph_state::CognitiveNodeFilter;
pub use engine::graph_state::CognitiveNodeOrder;

Modules§

action
CK-1.7: Action Schema Library + Candidate Generator
agenda
CK-2.1: Agenda / Open Loops Engine
analogy
CK-5.1 — Analogical Reasoning Engine.
attention
Working Set & Spreading Activation
belief
Belief Revision Engine — Bayesian Truth Maintenance
belief_network
CK-5.5 — Probabilistic Belief Network.
belief_query
Belief Query & Explanation — Pattern matching and provenance chains.
bench_utils
Shared utilities for benchmarks and profiling.
benchmark
Cognitive Kernel Benchmark Suite
benchmark_ck4
CK-4 Benchmark Suite — Advanced Reasoning Primitives + Meta-Cognition
calibration
Online Weight Learning + Confidence Calibration.
causal
CK-4.1 — Causal-Lite Inference Engine.
coherence
CK-4.3 — Coherence Monitor + Cognitive Health.
compression
conflict
Conflict detection and resolution.
consolidate
contradiction
Contradiction Detection — Finding and tracking conflicting beliefs.
counterfactual
CK-5.4 — Counterfactual Simulator.
encryption
engine
error
evaluator
CK-1.8: Utility Scoring + Bounded Forward Simulation
experimenter
Safe Self-Experimentation Engine — bounded A/B testing.
extractor
Cognitive Extractor Cascade — turns natural language into structured cognitive state updates.
flywheel
Autonomous Learning Flywheel — forms beliefs from observed correlations.
graph
Graph traversal utilities for entity-augmented recall.
graph_index
In-memory graph adjacency index for fast entity-augmented recall.
hawkes
CK-2.3: Routine Prediction via Hawkes Processes
hlc
Hybrid Logical Clock (HLC) for causal ordering of operations.
hnsw
Rust-native HNSW (Hierarchical Navigable Small World) vector index.
intent
CK-1.6: Intent Inference Engine
introspection
Self-Awareness Introspection Report.
metacognition
CK-4.4 — Meta-Cognition: Knowing What You Don’t Know.
narrative
CK-5.3 — Episodic Narrative Memory.
observer
Passive Event Observer — the foundation of autonomous learning.
patterns
Pattern mining across memories.
personality
personality_bias
CK-4.5 — Personality Bias Vectors.
perspective
CK-5.7 — Perspective Engine.
planner
CK-4.2 — Planning Graph / HTN-Lite Reasoning.
policy
CK-1.9: Policy Engine + Constraint Filtering
query_dsl
CK-4.6 — Cognitive Query DSL: Composable Reasoning Operators.
receptivity
CK-2.4: User Receptivity / Interruptibility Model
replay
CK-5.6 — Experience Replay / Dream Consolidation.
replication
Replication: oplog extraction, CRDT materialization, and operation application.
schema
schema_induction
CK-5.2 — Schema Induction Engine.
scoring
serde_helpers
skills
Autonomous Skill Acquisition — learns reusable multi-step capabilities from observed user action patterns.
state
Cognitive State Graph — Node types, edges, and attributes.
suggest
CK-1.10: suggest_next_step() — Full Pipeline Integration
surfacing
CK-2.6: Anticipatory Action Surfacing + Contextual Suppression
sync
Bidirectional sync protocol.
temporal
CK-2.2: Temporal Reasoning Primitives
tick
CK-2.5: cognitive_tick() — Background Cognition Heartbeat
triggers
types
vault
world_model
World Model Learning — action→outcome transition model.

Structs§

AbstainDecision
Decision about whether to proceed with autonomous action or defer.
AbstainReason
A reason contributing to an abstain decision.
ActionBandit
Beta-Bernoulli bandit for a single action kind.
ActionCandidate
A concrete action candidate: a schema bound to specific graph context.
ActionConfig
Configuration for candidate generation.
ActionProperties
Properties of an action that personality biases act upon.
ActionProposal
The action proposal — what the system suggests doing.
ActionRecord
What action was taken in an episode.
ActionTemplate
A parameterized action — generalized from concrete action instances.
ActivationContext
Context for evaluating activation conditions.
Agenda
The in-memory agenda — a collection of tracked items.
AgendaConfig
Configuration for the agenda engine.
AgendaId
Unique identifier for an agenda item.
AgendaItem
A single agenda item — an unresolved thing the system is tracking.
AnalogicalOpportunity
An automatically detected analogical opportunity.
AnalogicalQuery
Request to find analogies for a given subgraph pattern.
AnalogyMaintenanceReport
Decay report from store maintenance.
AnalogyStore
Indexed collection of discovered analogies, with domain-based lookup.
Anomaly
A detected anomaly from the cognitive tick’s monitoring.
AnticipatedEvent
An event type that the system anticipates happening soon.
ArcAlert
Alert about arc health issues.
ArcId
Unique identifier for a narrative arc.
AttentionBudgetConfig
Attention budget — limits how many suggestions per session.
AttentionConfig
Controls spreading activation behavior.
AutobiographicalTimeline
The user’s complete life timeline — all narrative arcs and milestones.
AutonomousBelief
An autonomously formed belief atom.
BPConfig
Configuration for belief propagation.
BPResult
Result of running belief propagation.
BanditRegistry
Collection of per-action-kind bandits.
BeliefConflict
A detected contradiction between beliefs.
BeliefContradiction
Two beliefs that contradict each other, both with non-trivial confidence.
BeliefDelta
A change to a belief discovered during replay.
BeliefExplanation
Full explanation of why a belief is held.
BeliefInventory
Aggregate statistics about the belief landscape.
BeliefNetwork
The belief network: a factor graph over belief nodes.
BeliefPattern
Pattern for querying beliefs.
BeliefRevisionConfig
Configuration for the belief revision engine.
BeliefStageBreakdown
Breakdown of beliefs by lifecycle stage.
BeliefStore
Persistent store of all autonomously formed beliefs.
BeliefVariable
A random variable in the belief network.
BetaPosterior
Beta distribution posterior for a variant.
BiasConfig
Weights and scaling for personality bias application.
BiasContribution
A single dimension’s contribution to the bias.
Blocker
Something that blocks plan execution.
BoundPrecondition
A precondition bound to the current state.
BurstConfig
Configuration for burst detection.
BurstResult
Burst detection result.
CachedSuggestion
A pre-computed proactive suggestion cached by the tick loop.
CalibrationMap
Isotonic calibration map over equal-width confidence bins.
CandidateGenerationResult
Result of candidate generation for a set of intents.
CandidateInference
A prediction derived from an analogy — transferring knowledge from the source domain to the target domain.
CausalConfig
Configuration for causal discovery and maintenance.
CausalDelta
A change to a causal edge discovered during replay.
CausalDiscoveryReport
Report from a discovery pass.
CausalEdge
A directed causal hypothesis: cause → effect.
CausalExplanation
Detailed explanation of a causal edge.
CausalStore
The causal knowledge base — all known causal hypotheses.
CausalSummary
Summary of the causal knowledge base.
CausalTrace
Full evidence chain for a causal hypothesis.
Chapter
A bounded segment within a narrative arc.
CircadianHistogram
24-hour histogram for circadian distribution of events.
CircadianProfile
Circadian (time-of-day) modulation of the base rate.
CognitiveAttrs
Universal attribute set carried by every cognitive node.
CognitiveEdge
A directed, typed, weighted edge in the cognitive state graph.
CognitiveNode
A complete cognitive node: identity + attributes + payload.
CognitivePipeline
A composable reasoning pipeline built from cognitive operators.
CognitiveStyle
How reasoning behaviour changes in a perspective.
CognitiveUpdate
A single extracted cognitive update.
CoherenceConfig
Configuration for coherence checking and enforcement.
CoherenceHistory
Rolling history of coherence scores.
CoherenceReport
Comprehensive cognitive health report.
CoherenceSnapshot
A timestamped coherence score.
ConfidenceReport
Detailed confidence and calibration report.
Conflict
A conflict between two memories.
ConflictResolutionResult
Result of a conflict resolution.
ConsolidationPreview
Dry run consolidation preview.
ConsolidationResult
Consolidation result (after consolidation runs).
ContextSnapshot
A snapshot of the user’s current context, used to build the feature vector.
ContradictionConfig
Configuration for contradiction detection.
ContradictionScanResult
Summary of contradiction detection results.
CorrectionResult
Result of a user-initiated correction.
CounterfactualConfig
Configuration for counterfactual simulation.
CounterfactualQuery
A counterfactual query: “what if…?”
CounterfactualResult
The full result of a counterfactual simulation.
CoverageGap
A specific area where the system lacks coverage.
CrossDomainLink
A link between memories in different domains discovered by cross-domain mining.
DeadlineAlert
A goal approaching or past its deadline.
DeadlineUrgencyConfig
Configuration for deadline urgency computation.
DecayedMemory
A decayed memory candidate from decay().
DecisionRecord
A decision record for regret analysis.
DependencyCycle
A cycle in the dependency graph.
DerivedSignals
Signals derived from recent observations — used by higher CK layers.
Discovery
A notable discovery — a belief that crossed the “interesting” threshold.
DiscoveryExplanation
Detailed explanation of a single discovery.
DiscoveryResult
Result of a skill discovery run.
DomainCount
Count of mentions within a domain.
DreamReport
Summary of a complete replay cycle (“dream report”).
Edge
An edge in the entity graph.
EdgeCorrespondence
An edge-level correspondence in a structural mapping.
EdgeWeightModifier
Adjust edge weights contextually.
EffectEstimate
The result of an effect estimation query.
EnforcementAction
A single enforcement action.
EnforcementReport
Actions taken during enforcement.
Entity
An entity in the knowledge graph.
EntityBridge
An entity that bridges multiple domains.
EntityProfile
Rich profile of an entity across time, domains, and sessions.
EpisodeData
A concrete (state, action, outcome) triple observed from the system. This is the raw input that drives schema induction.
EvaluatedAction
An action candidate with computed utility score.
EvaluationResult
Result of evaluating a set of action candidates.
EvaluatorConfig
Configuration for utility evaluation.
EventBuffer
Circular buffer for recent events. O(1) push, O(n) scan.
EventCounters
Per-event-kind counters with sliding window support.
EventFilter
Filter predicate for querying events.
EventPrediction
A prediction for when an event will next occur.
EventTypeModel
A complete Hawkes process model for one event type.
Evidence
A piece of evidence to be asserted into the belief system.
EvidenceContribution
Contribution of a single piece of evidence to the result.
EvidenceInput
New evidence to integrate into beliefs.
EvidenceResult
Result of asserting a piece of evidence.
EvolutionConfig
Configuration for personality evolution.
EwmaTracker
Exponentially Weighted Moving Average tracker.
ExpectedOutcome
Statistical outcome from applying a schema.
Experiment
A single self-experiment.
ExperimentRegistry
Registry of all experiments (active, concluded, aborted).
ExplanationTrace
A human-readable trace of the reasoning process.
ExtractionContext
Context provided to the extractor for disambiguation.
ExtractionResponse
Response from the extraction cascade.
ExtractorConfig
Configuration for the extraction cascade.
ExtractorSummary
Summary statistics for the extractor.
FactorId
Unique identifier for a factor in the belief network.
FlywheelConfig
Configuration for the autonomous belief formation pipeline.
FormationResult
Result of a belief formation run.
GoalConflict
Two active goals that work against each other.
HawkesParams
Parameters for a single Hawkes process model.
HawkesRegistry
Registry of Hawkes process models for multiple event types.
HawkesRegistryConfig
Configuration for the Hawkes registry.
InducedSchema
An abstract decision rule extracted from observations.
InducedSchemaStore
Collection of induced schemas with indexing.
InferenceQuery
A probabilistic query about the belief network.
InferenceResult
Result of an inference query.
IntentConfig
Configuration for intent inference.
IntentInferenceResult
Result of running intent inference on the cognitive graph.
InteractionRecord
Record of a single interaction for learning.
IntrospectionReport
Comprehensive introspection report — everything the system has learned.
LearnedPreferences
Learned preference adjustments from user feedback.
LearnedSkill
A learned multi-step skill with trigger context and reliability tracking.
LearnedWeights
Learned scoring weights stored per-database for adaptive recall.
LearningConfig
Configuration for the learning system.
LearningMilestone
A learning milestone — a significant event in the system’s cognitive growth.
LearningReport
Comprehensive report on the learning system’s state.
LearningState
Aggregate learning state — persisted as a single unit.
LlmExtractionRequest
Descriptor for a deferred LLM extraction request.
Memory
A memory record returned by get() and recall().
MemoryWithEmbedding
Internal struct with embedding data for clustering.
MetaCognitiveConfig
Configuration for meta-cognitive thresholds and weights.
MetaCognitiveHistory
Rolling history of meta-cognitive assessments.
MetaCognitiveReport
Comprehensive meta-cognitive assessment of reasoning quality.
MetaCognitiveSnapshot
A timestamped snapshot of meta-cognitive state.
Milestone
A significant life event that may span multiple arcs.
ModelSummary
Summary statistics for an event type model.
NarrativeArc
A coherent story thread spanning multiple episodes.
NarrativeEpisode
An episode to be classified into a narrative arc.
NarrativeResult
Result of a narrative query.
NetworkFactor
A factor (compatibility function) between variables.
NetworkHealth
Health diagnostic for the belief network.
NextStepRequest
Request for the suggest_next_step() pipeline.
NextStepResponse
Complete response from suggest_next_step().
NodeCorrespondence
A node-level correspondence in a structural mapping.
NodeDelta
A single node’s change between actual and counterfactual.
NodeId
Compact node identifier with embedded type tag.
NodeIdAllocator
Thread-safe sequence allocator for NodeIds. Maintains per-kind counters that persist across sessions via SQLite.
Observation
An observation of what actually happened (for regret analysis).
ObserverConfig
Configuration for the passive event observer.
ObserverState
The complete observer state — serializable for persistence.
ObserverSummary
Get a summary of observer activity.
OrphanedItem
An item with no supporting edges or parent context.
OutcomeData
Outcome data from an episode.
OutcomeDifference
Comparison between actual and counterfactual outcomes.
OutcomeDistribution
Dirichlet-Categorical distribution over action outcomes.
ParameterSlot
An abstract parameter that gets bound at execution time.
Pattern
A detected pattern across memories.
PatternConfig
Configuration for pattern mining.
PatternMiningResult
Result of pattern mining.
PeriodicityConfig
Configuration for periodicity detection.
PeriodicityResult
Full periodicity analysis result.
PersistedTrigger
A persisted trigger with lifecycle state.
PersonalityBiasResult
Result of applying personality bias to an action.
PersonalityBiasStore
Persistent personality state.
PersonalityBiasVector
An 8-dimensional personality bias vector.
PersonalityImpactReport
Human-readable report of how personality affected a decision.
PersonalityProfile
Aggregated personality profile across all traits.
PersonalityTrait
A single personality trait with its current score and derivation metadata.
Perspective
A reasoning context that reweights the cognitive graph.
PerspectiveConflict
A conflict between active perspectives.
PerspectiveId
Unique identifier for a perspective.
PerspectiveStack
The active perspective context (supports layering).
PerspectiveStore
All known perspectives.
PerspectiveTransition
A suggested perspective transition.
PipelineCandidateAction
A candidate action to evaluate.
PipelineContext
Context provided to the pipeline at construction time.
PipelineMetrics
Timing and count metrics for each pipeline stage.
PipelinePatterns
Pre-built pipeline patterns for common reasoning scenarios.
PipelineResult
Result of executing a cognitive pipeline.
Plan
A complete plan: an ordered sequence of steps to achieve a goal.
PlanProposal
Result of plan generation.
PlanScore
Composite plan score.
PlanStep
A single step in a plan.
PlanStore
Stores active plans for goals.
PlannerConfig
Configuration for the planner.
PolicyConfig
Configuration for the policy engine.
PolicyConstraint
A policy constraint applied to filter candidates.
PolicyContext
Snapshot of the user’s current context for policy decisions.
PolicyResult
Complete policy engine result.
PredictedEffect
A predicted downstream effect.
ProactiveSuggestion
A proactive suggestion ready to present to the user.
ProjectedFact
A projected fact derived from an analogy — a prediction about the target domain based on knowledge in the source domain.
QuietHoursConfig
Quiet hours configuration — time windows where interruptions are blocked.
ReasoningHealthReport
Overall reasoning health report.
ReasoningTrace
The full reasoning trace for a policy decision.
RecallQuery
Builder for composable recall queries.
RecallResponse
Response from recall with confidence and hints for interactive retrieval.
RecallResult
A recall result with scoring information.
RecencyConfig
Configuration for recency-weighted relevance scoring.
ReceptivityEstimate
Full receptivity estimate with explainable factors.
ReceptivityFactor
A single factor contributing to the receptivity score.
ReceptivityModel
Online logistic regression model for receptivity prediction.
RecordInput
Input for batch record operations.
RefinementHint
A hint for refining a query when confidence is low.
RegretReport
Aggregated regret analysis across multiple decisions.
RelationshipDepth
Rich interaction metrics for an entity, measuring depth of knowledge. This goes beyond simple mention counts to capture how deeply the system knows about an entity across sessions, domains, and time.
ReliabilityRegistry
Collection of source reliability trackers.
ReplayBudget
Controls how much replay happens per cycle.
ReplayBuffer
Prioritized collection of replayable episodes.
ReplayEngine
The replay engine: manages the experience replay process.
ReplayEntry
A single episode queued for replay.
ReplayOutcome
Result of replaying a single episode.
ReplayStats
Replay statistics.
ReplaySummary
Summary of the replay engine state.
RetrievalSummary
Summary of how retrieval was performed.
RevisionSummary
Summary of a batch belief revision pass.
SalienceOverride
Boost or dampen a node’s salience.
SchemaContext
A snapshot of the current cognitive context for schema matching.
SchemaId
Unique identifier for an induced schema.
SchemaMaintenanceReport
Report from schema maintenance.
ScoreBreakdown
Score breakdown for a recall result.
ScoreContributions
Weighted contributions of each signal (signal_value * weight).
ScoredIntent
A scored intent hypothesis ready for ranking.
ScoringRow
Lightweight scoring fields cached in memory for fast recall scoring. These are the only fields needed to compute composite_score() during recall.
SeasonalHistogram
A histogram that bins events by time-of-day or day-of-week.
SelectedAction
A selected action with adjusted score and trace.
SensitivityEntry
Sensitivity analysis result: which factors matter most.
Session
A session tracks a conversation or interaction period.
SessionSummary
Summary returned when ending a session.
SignalDetail
A single meta-cognitive signal with its value and weight.
SimulatedStep
One step in the causal simulation chain.
SkillConfig
Configuration for the skill acquisition engine.
SkillMatch
A scored skill match for the current context.
SkillRegistry
The skill registry — stores all learned skills.
SkillStep
A single step within a learned multi-step skill.
SkillSummary
Summary statistics for the skill registry.
SkillTrigger
Context conditions under which a skill should be triggered.
SourceReliability
Reliability tracker for a single evidence source.
StaleNode
A node that is “active” but hasn’t been updated recently.
StateFeatures
Compact discretized state features.
StateSnapshot
Snapshot of the predicted end state after simulation.
Stats
Engine statistics.
StepResult
Result of a single operator execution.
StructuralMapping
Alignment between two subgraphs — the core data structure of analogical reasoning. Maps nodes and edges from a source domain to a target domain, measuring structural similarity.
SubgraphGroup
A group of nodes+edges forming a candidate target subgraph.
SuppressedItem
A suppressed item — for debugging and learning why things didn’t surface.
SurfaceRateLimiter
Tracks recent surfacing events for rate limiting.
SurfacingConfig
Configuration for the surfacing pipeline.
SurfacingPreferences
Persistent surfacing preferences learned from user behavior.
SurfacingResult
Result of the surfacing pipeline.
SystemEvent
A system-wide event observed by the passive observer.
TemplateConstraint
Constraints on parameter values.
TemplateStore
Template store — holds all learned extraction templates.
TemporalMotif
A discovered temporal motif (recurring sequence of event types).
TemporalOrder
Result of topological ordering of temporal events.
TemporalRelevanceConfig
Configuration for composite temporal relevance scoring.
ThinkConfig
Configuration for the think() cognition loop.
ThinkResult
Result of a think() pass.
TickConfig
Configuration for the cognitive tick loop.
TickReport
Report produced by a single cognitive tick.
TickResult
Result of a tick operation — what changed.
TickState
Persistent state for the cognitive tick loop.
TimeInterval
A closed time interval [start, end] in seconds (unix timestamp).
TransferredStrategy
A transferred action schema — an action template adapted from a source domain to a target domain using a structural mapping.
TransitionModel
The world model: maps (state, action) → outcome distribution.
Trigger
A proactive trigger.
TurningPoint
A moment of significant change in an arc.
UtilityWeights
Learned weights for the utility evaluator.
VariableId
Unique identifier for a variable in the belief network.
WeightSnapshot
Weight snapshot for external consumption.
WhatIfResult
Complete what-if analysis result.
WorkingSet
The bounded in-memory cognitive working set.
WorldModelSummary
Summary of the world model’s learned knowledge.

Enums§

AbstainAction
What action to take when the system is uncertain.
ActivationCondition
When a perspective should auto-activate.
ActivityState
The user’s current activity state, used to estimate interruption cost.
AgendaKind
What kind of open loop this represents.
AgendaStatus
Lifecycle status of an agenda item.
AnalogyScope
Scope for searching analogies.
AnomalyKind
Types of anomalies the tick loop can detect.
ArcAlertType
Types of arc health alerts.
ArcStatus
Lifecycle status of a narrative arc.
ArcTheme
The thematic category of a narrative arc.
BeliefCategory
Category of autonomously formed belief.
BeliefEvidence
Category-specific evidence backing a belief.
BeliefOrder
Sort order for belief queries.
BeliefStage
Confidence lifecycle stage.
BlockerKind
Categories of blockers.
BondLevel
Bond level affects how much personality is expressed.
CausalEvidence
A single piece of causal evidence.
CausalNode
Identifies a cause or effect node.
CausalStage
Lifecycle stage of a causal hypothesis.
ChapterType
The type of a chapter within a narrative arc.
CognitiveEdgeKind
Typed edges in the cognitive state graph. Each edge kind has semantic meaning that affects spreading activation and reasoning. Weights are always in [-1.0, 1.0].
CognitiveOperator
A single cognitive operator in the pipeline.
ConflictDetectionMethod
How a contradiction was detected.
ConflictType
Type of conflict between perspectives.
ConflictType
The type of semantic conflict between two memories.
ConstraintExpr
Constraint expressions for template parameters.
ConstraintKind
Types of policy constraints.
CounterfactualType
The type of counterfactual question being asked.
CoverageGapKind
Types of coverage gaps.
DeltaDirection
Direction of a counterfactual change.
DirectionChange
Direction of change at a turning point.
DiscoveryMethod
How the causal hypothesis was discovered.
Distribution
Probability distribution representation.
EdgeRelation
Edge relation type for network construction.
EnforcementKind
Types of enforcement actions.
EventKind
Discriminator for event types — used as hash map keys and filter predicates.
EvidenceQuality
Quality tier of causal evidence.
EvidenceSource
Evidence source types for reliability tracking.
ExecutionMode
How deeply to analyze before responding.
ExperimentStatus
Status of an experiment.
ExperimentVariable
What parameter is being varied.
ExtractorTier
How the extraction was derived.
FactorType
The semantic type of a factor (derived from cognitive edge kind).
InferenceType
Type of probabilistic query.
IntentSource
What signal source generated this intent hypothesis.
InteractionOutcome
The outcome of a suggestion/action offered to the user.
IntervalRelation
Allen’s 13 interval relations — the complete set for temporal reasoning.
Intervention
An intervention in the structural causal model.
MilestoneKind
Types of learning milestones.
NarrativeQuery
A query against the autobiographical timeline.
NodeKind
The 14 cognitive node types that make up the state graph.
NodePayload
Kind-specific data carried by each cognitive node. This is the union of all node-type-specific fields.
NotificationMode
Notification mode — user’s current notification preference.
ParamType
Types of parameters in an action template.
PersonalityPreset
Named personality presets.
PerspectiveType
Category of perspective.
PipelineExecutionMode
How the pipeline is executed.
PipelineStatus
Status of the pipeline execution.
PolicyDecision
The outcome of policy selection.
PotentialFunction
Potential function — how variables interact in a factor.
ProjectionHorizon
How far forward to project.
Provenance
How a cognitive node was created — critical for trust and revision.
ResolutionStrategy
Suggested way to resolve a contradiction.
SafetyBound
Safety bound — when violated, experiment aborts.
SalienceTarget
What to target with a salience override.
SamplingStrategy
Sampling strategy for the replay buffer.
SchemaCondition
An abstract precondition — not tied to specific entities.
SchemaDirection
Direction of an edge in a precondition.
SerializableOpTemplate
Serializable version of OpTemplate (for template persistence).
SignalStatus
Health status of a meta-cognitive signal.
SkillOrigin
How the skill was discovered.
SkillStage
Lifecycle stage of a learned skill.
StepDerivation
How a plan step was derived.
StepOutput
Operator-specific output variants.
SuggestionOutcome
What the user did with a suggestion.
SuppressionCause
Why an item was suppressed (not surfaced).
SurfaceMode
How prominently to surface an action to the user.
SurfaceOutcome
Outcome of a surfaced suggestion — used to update surfacing thresholds.
SurfaceReason
Why an item is being surfaced.
SystemEventData
Event payload — typed variants for each observable event.
TemporalFocus
Temporal focus mode within a cognitive style.
ThresholdDirection
Direction of a threshold crossing.
TickPhase
Which phase of the tick loop was executed.
TransferType
How knowledge was transferred via an analogy.
TrialOutcome
Outcome of a single experiment trial.
TriggerType
Trigger type classification.
UpdateOp
The operation to perform on the cognitive graph.
UrgencyFn
Parameterized urgency function that computes urgency from time.
VariantValue
A specific value for an experiment variable.
WorldActionKind
High-level action kind (what the system did).
WorldActionOutcome
Observed outcome of an action.
YantrikDbError

Traits§

Embedder
Trait for converting text to embedding vectors. Implementations can use any embedding model (sentence-transformers, candle, etc.).
PipelineExecutor
Trait implemented by the engine layer to execute individual operators.

Functions§

check_all_triggers
Run all trigger checks and return a unified, priority-sorted list.
check_consolidation_triggers
Trigger when there are enough active memories that consolidation might help.
check_decay_triggers
Find important memories that are decaying significantly.
consolidate
Run the full consolidation pipeline.
create_conflict
Create a conflict record and log it to the oplog for replication.
derive_personality
Derive personality traits from memory signals and update the DB via EMA blending. Returns the updated personality profile.
detect_edge_conflicts
Detect edge-based contradictions for a newly materialized edge. Called from materialize_relate in replication.rs during sync.
find_consolidation_candidates
Find clusters of memories that are candidates for consolidation.
get_personality
Read the current personality profile from the DB without recomputing.
mine_patterns
Run all pattern mining algorithms.
scan_conflicts
Full-database conflict scan. Finds all edge-based contradictions and concurrent consolidation conflicts.
set_personality_trait
Manually set a personality trait score (for testing/override).

Type Aliases§

ExperimentId
Unique experiment identifier.
SkillId
Unique identifier for a learned skill.