Crate tpt_eve_core 

Shared types for the TPT Eve cognitive architecture.

Defines the core vocabulary used across all Eve crates:

• Value — a typed atom (Symbol or Text).
• Fact — a subject–predicate–object triple with provenance and confidence.
• Pattern — a triple where any slot may be bound or variable.
• EveError — the shared error type.
• CausalRelationDraft / CausalRelationKind — Phase 2 causal-relation types defined here so tpt-eve-neural’s trait surface matches the spec from day one.

Structs

CausalRelationDraft

A causal relationship as extracted from text by the Neural Pattern Layer, before it has been validated or integrated into a causal graph.

Fact

A subject-predicate-object triple held in working memory.

Pattern

A triple pattern used for querying working memory. Any combination of slots may be bound or variable.

Enums

CausalRelationKind

EveError

Shared error type for all eve-* crates. Kept dependency-free (no eve-dsl/eve-symbolic types referenced directly) so eve-core stays at the bottom of the workspace dependency graph.

FactSource

Where a fact came from, for provenance and confidence tracking.

PatternTerm

One slot of a Pattern: either bound to a concrete value or an unbound variable to be solved for during unification.

Value

A single term in a fact triple. Symbol is an identifier-like token (e.g. TypeScript, has); Text is a free-form quoted phrase (e.g. "programming language").