Codebase Contextifier provides a robust, Docker-based Model Context Protocol (MCP) server for comprehensive code understanding. It leverages AST-aware chunking to preserve semantic integrity, tracks code relationships (calls, imports, inheritance) using a Neo4j graph database, and enables powerful semantic search across multiple repositories via a Qdrant vector database. With features like incremental indexing, real-time file monitoring, and local LLM support via Ollama, it offers an efficient and private solution for developers to navigate, analyze, and query their codebases.
Key Features
010 GitHub stars
02Relationship Tracking via Neo4j graph database
03Multi-Repository Semantic Search powered by Qdrant
04AST-Aware Chunking for semantic integrity
05Real-Time Updates with file system watcher
06Incremental Indexing with Merkle tree-based change detection
Use Cases
01Indexing and searching large, polyglot codebases for semantic content
02Providing accurate, context-rich information for local LLM-powered coding assistants
03Tracking function usages and dependencies across multiple projects