Skip to main content

​
Pipeline

Source files
     │
     ▼
 [Chunker]  ──  tree-sitter AST → semantic chunks (functions, classes)
     │           fallback: sliding-window for unsupported file types
     │
     ▼
 [Embedder] ──  fastembed ONNX (default) or sentence-transformers (torch)
     │           all-MiniLM-L6-v2-code-search-512 (384-dim)
     │
     ▼
 [Store]    ──  LanceDB vector store (~/.vecgrep/<project_hash>/index.db)
     │           IVF-PQ ANN index for sub-linear search
     │           file_stats table for O(files) change detection
     │
     ▼
 [Server]   ──  MCP server exposing index_codebase, search_code, get_index_status

​
Index location

Each project gets its own isolated index: 
~/.vecgrep/<sha256-of-absolute-project-path>/index.db
Delete the directory to wipe the index and start fresh.

​
Change detection

VecGrep tracks mtime and file size in a separate file_stats table. On re-index, only files whose mtime or size has changed are re-embedded — unchanged files are skipped in O(1) per file.