What is Dextree
Dextree turns your repository into an interactive knowledge graph. Every symbol, call, import, and dependency is indexed into an embedded graph database (DuckDB + DuckPGQ) and made explorable through a WebGL-rendered visualization directly inside VS Code.
Unlike competitors that live outside the editor, Dextree fuses the graph with everything VS Code already knows — LSP type information, live diagnostics, git history, and test linkage — to give you signals no headless tool can produce.
Why a code graph at all?
Most code intelligence tools answer "what does this symbol do?" Dextree answers a different question: what depends on it, and what would break if I change it?
That sounds obvious. It isn't — because the answer requires:
- A structural graph of every file in the repo (Tree-sitter pass)
- A semantic graph with type-accurate edges (LSP pass)
- Live context from your editor: diagnostics, blame, test discovery
- A query engine fast enough to answer in milliseconds
Everyone has one or two. Dextree has all four, fused into one DuckDB file. No server, no daemon, no tsserver of doom — just an embedded graph that knows your code the way you do.
What you'll find in this guide
- Installation — set up the extension and run your first index.
- Two-pass indexing — how Tree-sitter and LSP cooperate to keep the graph both fast and accurate.
- Shared graph — why every surface (webview, exporters, MCP, CLI, Alfred) reads the same DuckDB substrate.
- VS Code data fusion — the architectural moat that lets us answer questions headless competitors can't.
Status
Dextree is pre-alpha. v0.1 is in active development and not yet on the VS Code Marketplace. Follow the roadmap for the slice-by-slice plan, or watch the repo for the release notification.
TIP
The graph is the product. Visualizations, exports, MCP, and Alfred are surfaces over the same indexed substrate — change the schema once, every surface benefits.