Codebase-Memory: Tree-Sitter-Based Knowledge Graphs for LLM Code Exploration via MCP

TL;DR

Codebase-Memory is an open-source system that builds a persistent, language-agnostic knowledge graph from codebases using Tree-Sitter, improving LLM code exploration efficiency and accuracy.

Contribution

It introduces a novel Tree-Sitter-based knowledge graph construction method via MCP, enabling efficient, multi-language code understanding for LLM agents.

Findings

Achieves 83% answer quality with fewer tokens and tool calls.

Matches or exceeds explorer on hub detection and caller ranking in most languages.

Supports 66 languages through a multi-phase, parallel pipeline.

Abstract

Large Language Model (LLM) coding agents typically explore codebases through repeated file-reading and grep-searching, consuming thousands of tokens per query without structural understanding. We present Codebase-Memory, an open-source system that constructs a persistent, Tree-Sitter-based knowledge graph via the Model Context Protocol (MCP), parsing 66 languages through a multi-phase pipeline with parallel worker pools, call-graph traversal, impact analysis, and community discovery. Evaluated across 31 real-world repositories, Codebase-Memory achieves 83% answer quality versus 92% for a file-exploration agent, at ten times fewer tokens and 2.1 times fewer tool calls. For graph-native queries such as hub detection and caller ranking, it matches or exceeds the explorer on 19 of 31 languages.