query-graph

You can query a knowledge graph to find related source code and specifications using natural language or Cypher queries. This skill surfaces relevant entities, files, and design docs by searching a prebuilt RyuGraph knowledge graph.

Yes, you can use natural language instead of Cypher to search code and specs in a graph database. This skill supports querying the RyuGraph knowledge graph using either natural language or Cypher to surface related entities.

To map a ubiquitous domain term to specific source code files and design docs, query the knowledge graph with the term. The skill returns related entities, their definitions, and associated specifications for quick understanding.

Yes, you need a prebuilt RyuGraph knowledge graph located at knowledge.ryugraph before using this code search skill. You also need Python 3.9+ and the ryugraph package installed in your environment.

The best way to locate design documentation for a specific software entity is querying the graph with a Cypher statement like 'MATCH (e:Entity)-[:HAS_TERM]->(t:UbiquitousTerm) RETURN e.name'. This surfaces the entity and its related design docs.

Graph-based code search requires a prebuilt knowledge graph because the skill queries existing nodes and relationships to surface related code and specifications. It does not build the graph; it assumes one exists at knowledge.ryugraph.