Understand any codebase instantly. System intelligence for codebases, built for humans and AI.
TypeScript aiai-memoryclaude-codeclicode-analysiscode-mappingcode-memorycodebase-understandingcodexdeveloper-toolsllmopenclawpersistent-memoryprogram-analysissoftware-architecturesystem-intelligencesystem-mapping
feat: wire nodeKindMap into RelevanceScorer for symbol preference
riley022701a2ee236949
diff --git a/memory-layer/src/main/scala/ix/memory/context/ContextService.scala b/memory-layer/src/main/scala/ix/memory/context/ContextService.scalaindex d79c9664..81609502 100644--- a/memory-layer/src/main/scala/ix/memory/context/ContextService.scala+++ b/memory-layer/src/main/scala/ix/memory/context/ContextService.scala@@ -66,8 +66,9 @@ class ContextService())}- // 7. Relevance score: weight claims by hop distance from seeds, boost exact matches- relevant = RelevanceScorer.scoreWithTerms(scored, seeds.map(_.id).toSet, expanded.edges, terms)+ // 7. Relevance score: weight claims by hop distance from seeds, boost exact matches, prefer symbols+ nodeKindMap: Map[NodeId, NodeKind] = (seeds ++ expanded.nodes).map(n => n.id -> n.kind).toMap+ relevant = RelevanceScorer.scoreWithTerms(scored, seeds.map(_.id).toSet, expanded.edges, terms, nodeKindMap)// 8. Rank claims by finalScore (relevance x confidence) descendingranked = ContextRanker.rank(relevant)