What Enliterator Does

Enliterator transforms datasets into Enliterated Knowledge Navigators (EKNs)—graph‑constrained collaborators whose inference is bounded by structured knowledge graphs, ensuring reliable, grounded responses that you can trust, explore, and expand over time.

Pipeline (Stages 1–8)

  • Intake: Discover, hash, and partition files into IngestItems.
  • Rights & Provenance: Attach consent, license, ownership, publishability, and training eligibility.
  • Lexicon Bootstrap: Extract canonical terms and surface forms using OpenAI Structured Outputs.
  • Pool Filling: Populate the Ten Pool Canon (Idea, Manifest, Experience, Relational, Evolutionary, Practical, Emanation, …).
  • Graph Assembly: Build a Neo4j knowledge graph with constraints, indexes, and rich repr_text.
  • Representations & Retrieval: Generate embeddings directly in Neo4j (GenAI), enable semantic/hybrid search.
  • Literacy Scoring: Assess coverage, maturity, and gaps with actionable recommendations.
  • Deliverables: Export graph dumps, prompt packs, evaluation bundles, and multi‑format outputs.

Graph‑Constrained Collaborators (EKNs)

Each EKN maintains identity, accumulates knowledge across batches, and provides graph‑bounded inference through a conversational interface. EKNs can expose their knowledge graphs via MCP for nested collaboration, enabling sophisticated multi‑domain reasoning while preserving inference reliability. The Meta‑Enliterator is an EKN that helps create other EKNs.

Knowledge Structure: Ten Pool Canon + Domain Pools

Enliterator creates structured knowledge boundaries through the Ten Pool Canon—a universal schema that models how meaning flows between different types of knowledge. Each pool captures a fundamental aspect of understanding, enabling graph‑constrained inference.

Core Pools (1-10): Universal Knowledge Types

1
Idea

Principles, theories, design rationales—the why

2
Manifest

Concrete instances, artifacts, projects—the what

3
Experience

Lived outcomes, testimonials, observations

4
Relational

Connections, networks, collaboration patterns

5
Evolutionary

Change over time, versions, development

6
Practical

How‑to knowledge, guides, tacit expertise

7
Emanation

Ripple effects, downstream influence, adoption

8
Provenance & Rights

Source, attribution, consent, licensing

9
Lexicon & Ontology

Definitions, canonical terms, schema

10
Intent & Task

User goals, presentation preferences, success criteria

Domain Pools (+5): Specialized Extensions

11
Actor & Role

People, organizations, permissions, governance

12
Spatial

Places, regions, geometries, location hierarchies

13
Evidence & Observation

Primary data, measurements, raw findings

14
Risk & Governance

Hazards, mitigations, compliance, safety gates

15
Method & Model

Methodologies, evaluation patterns, reproducibility

Constraint Mechanism: Domain pools are added only when concepts have independent lifecycles and many‑to‑many relations, preventing schema bloat while ensuring comprehensive coverage.

How Pools Enable Graph‑Bounded Inference

  • • Structured boundaries: Each pool defines valid entity types and relationships
  • • Closed relation vocabulary: Connections use a formal verb glossary (embodies, elicits, validates, etc.)
  • • Provenance tracking: Every entity traces back to source documents with rights metadata
  • • Inference constraints: Responses can only reference entities and paths that exist in the graph
  • • Canonical normalization: Lexicon pool ensures consistent terminology across the knowledge base

What makes it different?

  • Graph‑bounded inference: Responses constrained by actual knowledge, not probabilistic hallucination.
  • Rights‑aware collaboration: From ingest to answer synthesis with full provenance tracking.
  • MCP‑enabled nesting: EKNs can collaborate via shared knowledge graphs while maintaining constraints.
  • Transparent modeling: Ten Pool Canon provides structured knowledge boundaries.
  • Neo4j + vector hybrid: Graph structure constrains while vectors enable semantic search.
  • Deterministic extraction: Structured Outputs ensure reproducible knowledge representation.

Highlights

  • Zero‑touch pipeline from intake to deliverables
  • Rights-aware processing with provenance throughout
  • Ten Pool Canon entity modeling (Idea, Manifest, Experience, …)
  • Neo4j knowledge graph with vector search (GenAI)
  • OpenAI-powered extraction using Structured Outputs
  • Enliteracy scoring, gaps, and maturity assessment
  • Autogenerated deliverables: exports, prompt packs, evaluation
  • Knowledge Navigators (EKNs) for conversational exploration

Ready to Explore?

Explore Graph‑Constrained Collaboration 🚀 See Dynamic UI Vision