ontology-enrich

name	ontology-enrich
description	Validate and enrich the Logseq mainKnowledgeGraph ontology and generate WebVOWL-ready TTL. Use when fixing source-domain prefixes (ai/bc/mv/rb/tc/ngm), regenerating output/ontology.ttl, debugging WebVOWL 'prefix not bound' or bad-syntax parse errors, checking orphan is-subclass-of targets, or making pages meet VisionClaw github_sync field requirements.
version	2.0.0
category	ontology
layer	1
dependencies	["ontology-core"]
tags	["ontology","validation","enrichment","owl2","ttl","webvowl","perplexity"]

Ontology Enrich Skill

When Not To Use

• For creating new ontology schemas from scratch -- use ontology-core instead
• For general data validation unrelated to Logseq ontology -- use standard validation tools
• For VisionClaw graph rendering -- this handles data, not display

Purpose

Validated enrichment and TTL generation for mainKnowledgeGraph corpus with VisionClaw/WebVOWL compatibility.

Key Workflows

1. Validate source-domain Values

CRITICAL: Only 6 valid prefixes exist:

# Find invalid source-domain values
grep -rhn "source-domain::" mainKnowledgeGraph/pages/*.md | \
  sed 's/.*source-domain::\s*//' | sort | uniq -c | sort -rn

# Valid: ai, bc, mv, rb, tc, ngm
# INVALID: blockchain, metaverse, telecollaboration, data, etc.

Fix invalid values:

# blockchain -> bc
grep -rln "source-domain:: blockchain" mainKnowledgeGraph/pages/*.md | \
  xargs -I {} sed -i 's/source-domain:: blockchain/source-domain:: bc/g' {}

# metaverse -> mv
grep -rln "source-domain:: metaverse" mainKnowledgeGraph/pages/*.md | \
  xargs -I {} sed -i 's/source-domain:: metaverse/source-domain:: mv/g' {}

# telecollaboration -> tc
grep -rln "source-domain:: telecollaboration" mainKnowledgeGraph/pages/*.md | \
  xargs -I {} sed -i 's/source-domain:: telecollaboration/source-domain:: tc/g' {}

2. Generate TTL

python3 Ontology-Tools/tools/converters/convert-to-turtle.py \
  mainKnowledgeGraph/pages/ \
  output/ontology.ttl

Output: Single output/ontology.ttl (git provides versioning)

3. Validate TTL for WebVOWL

# Check @prefix comes first (REQUIRED for format detection)
head -5 output/ontology.ttl | grep "@prefix"

# Check for unbound prefixes
grep -n "blockchain:\|metaverse:\|telecollaboration:\|data:" output/ontology.ttl

# Check for illegal characters in local names
grep -n "&\|(\|)" output/ontology.ttl | grep -v "rdfs:comment\|rdfs:label"

4. Enrich with Perplexity (Optional)

# Use perplexity-research skill for stub pages
# Query format for UK English context:
python -c "
from perplexity_client import query
result = query('''
Context: UK-based AI/Blockchain ontology documentation.
Task: Define {TERM} with technical accuracy.
Include: key components, relationships, UK context.
Output: 2-3 sentence definition suitable for ontology.
''')
"

Common Issues and Fixes

WebVOWL Parser Errors

Error	Root Cause	Fix
Prefix "X:" not bound	Invalid source-domain value	Use valid 2-letter prefix
Prefix ":" not bound	Bare colon in property decls	Use ngm: prefix for properties
Bad syntax (']' expected)	& in WikiLink target	sanitize_local_name()
unexpected token '#'	Comments before @prefix	@prefix MUST be line 1
Encountered '['	WikiLinks in definition	sanitize_literal()

Fixing Malformed Pages

# Find pages with all fields on one line
grep -l "ontology:: true.*term-id::" mainKnowledgeGraph/pages/*.md

# Find pages with & in relationships
grep -rn "enables.*&\|requires.*&\|has-part.*&" mainKnowledgeGraph/pages/*.md

Relationship Best Practices

is-subclass-of targets must exist

# Check for orphan relationships
grep -rn "is-subclass-of::" mainKnowledgeGraph/pages/*.md | \
  sed 's/.*\[\[\([^]]*\)\]\].*/\1/' | sort | uniq | \
  while read term; do
    if ! ls "mainKnowledgeGraph/pages/$term.md" 2>/dev/null; then
      echo "ORPHAN: $term"
    fi
  done

Cross-cutting with belongsToDomain

- belongsToDomain:: [[AIApplicationsDomain]], [[DisruptiveTechDomain]]

Use for cross-domain classification without changing source-domain.

Deployment Workflow

# .github/workflows/publish.yml
- name: Generate ontology files
  run: |
    if [ -f "output/ontology.ttl" ]; then
      cp output/ontology.ttl /tmp/narrativegoldmine-ontology.ttl
      echo "Using output/ontology.ttl"
    else
      echo "ERROR: output/ontology.ttl not found"
      exit 1
    fi

Integration with VisionClaw

VisionClaw's github_sync_service.rs expects:

Field	Required	Format
### OntologyBlock	Yes	Markdown heading
term-id::	Yes	DOMAIN-NNNN
owl:class::	Yes	domain:ClassName
source-domain::	Yes	2-letter code
public-access::	Yes	true or false
definition::	Yes	Clean text (no WikiLinks)

Quick Validation Script

#!/bin/bash
# validate-ontology.sh

echo "=== Checking source-domain values ==="
grep -rhn "source-domain::" mainKnowledgeGraph/pages/*.md | \
  sed 's/.*source-domain::\s*//' | sort | uniq -c | sort -rn

echo "=== Regenerating TTL ==="
python3 Ontology-Tools/tools/converters/convert-to-turtle.py \
  mainKnowledgeGraph/pages/ output/ontology.ttl

echo "=== Checking for unbound prefixes ==="
grep -c "blockchain:\|metaverse:\|data:" output/ontology.ttl && \
  echo "ERROR: Unbound prefixes found" || echo "OK: No unbound prefixes"

echo "=== Verifying @prefix first ==="
head -1 output/ontology.ttl | grep -q "@prefix" && \
  echo "OK: @prefix is first" || echo "ERROR: @prefix not first"

References

• Converter: Ontology-Tools/tools/converters/convert-to-turtle.py
• Workflow: .github/workflows/publish.yml
• TTL Output: output/ontology.ttl
• Domain pages: mainKnowledgeGraph/pages/*Domain.md