VectorCode

VectorCode is a code repository indexing tool designed to enhance coding LLM prompts by providing relevant code context. It includes a corresponding Neovim plugin. Currently in beta, it offers basic retrieval and embedding functionalities using Chromadb as the vector database backend, with SentenceTransformer as the default embedding engine.

VectorCode improves LLM code completion by providing project context, enabling more accurate suggestions compared to context-free completions. This approach is similar to that used in Continue.dev.

The documentation covers CLI usage, Neovim plugin setup, and contribution guidelines. Key features include querying by excluded paths, chunking support with metadata and configurable chunk size, smarter chunking using py-tree-sitter, Neovim Lua API with caching, persistent client, .gitignore support, and project-root anchors. Future plans involve remote Chromadb support, file management within collections, and joint search capabilities.