# How to Build a Local 3D Codebase Knowledge Graph and Sync LLM Context Offline

> Source: <https://dev.to/sreekanthap/how-to-build-a-local-3d-codebase-knowledge-graph-and-sync-llm-context-offline-1cdi>
> Published: 2026-06-16 19:42:09+00:00

Introduction: Why context windows are expensive and why static folder indexing fails to capture structural relationships (like function calls or imports).

The Solution: Introducing Graphify Companion.

How the Architecture Works: Detail the combination of Node.js for the MCP Server Transport and Python for AST/graph parsing.

Step-by-Step Setup:

Cloning the repo([https://github.com/sreekanthap89/Graphify-Companion1](https://github.com/sreekanthap89/Graphify-Companion1)).

Running install.ps1 to configure Python .venv and Git Hooks.

Starting LM Studio's local server.

Copy-pasting the JSON configuration into LM Studio.

Prompting Guide: Teach readers how to write prompts that force the AI to use graph tools (query_graph vs. read_file).

Conclusion & Source: Invite readers to star the GitHub repository.