libgraph Skill

When to Use

• Building and querying knowledge graphs
• Storing RDF triples for semantic data
• Executing subject-predicate-object pattern queries
• Working with linked data and ontologies

Key Concepts

GraphIndex: Storage-backed index for RDF triples with pattern-based querying.

PREFIXES: Standard namespace prefixes for common vocabularies (schema.org, RDF, RDFS, FOAF).

Triple patterns: Query graphs using subject-predicate-object patterns with wildcards.

Usage Patterns

Pattern 1: Query graph triples

import { createGraphIndex, PREFIXES } from "@copilot-ld/libgraph";

const index = await createGraphIndex(storage, "knowledge");
const results = await index.query(
  `${PREFIXES.schema}Person`,
  `${PREFIXES.schema}name`,
  "?",
);

Pattern 2: Parse triple query string

import { parseTripleQuery } from "@copilot-ld/libgraph";

const { subject, predicate, object } = parseTripleQuery("schema:Person ? ?");

Integration

Used by Graph service for knowledge queries. Processes ontology.ttl files.