Skip to content
GitLab Knowledge Graph
GitLab Knowledge Graph
GitLab Knowledge Graph Logo GitLab Knowledge Graph Logo

GitLab Knowledge Graph (gkg)

Create structured, queryable representations of code repositories to power AI features and enhance developer productivity.
Get Started View on GitLab

Why GKG?

Section titled “Why GKG?”

Universal Call Graph

Native parsing

Support for Rust, TypeScript, Python, Ruby, and more. Extensible architecture powered by gitlab-code-parser means new languages can be added with ease.

Lightning-Fast Queries

Kuzu Graph Database

Purpose-built for complex relationship queries. Find dependencies, trace function calls, and explore architecture at any scale.

Developer-First Design

Many Ways to Access

Powerful CLI for automation, IDE integration, and beautiful web UI for exploration. Works with all major OS.

AI-Ready Integration

MCP Protocol Support

Enables AI tools to understand your codebase structure and provide context-aware assistance.

Get Started in Minutes

Section titled “Get Started in Minutes”

  1. Install GitLab Knowledge Graph

    One-line installation:

    Terminal window
    curl -fsSL https://gitlab.com/gitlab-org/rust/knowledge-graph/-/raw/main/install.sh | bash

    ⚠️ Note: Ensure $HOME/.local/bin is in your $PATH environment variable. Add this line to your shell configuration file (e.g., ~/.bashrc, ~/.zshrc):

    Terminal window
    export PATH="$HOME/.local/bin:$PATH"

  2. Index your first repository

    Navigate to your project and create the knowledge graph:

    Terminal window
    gkg index

  3. Explore with the web interface

    Start the server and open your browser:

    Terminal window
    gkg server start

Use Cases

Section titled “Use Cases”
Codebase RAG Enable context-aware code suggestions and analysis with MCP integration
Code Navigation Find all references, dependencies, and relationships across your codebase
Impact Analysis Understand how changes affect other parts of your system before you make them
Architecture Visualization Generate interactive graphs of system structure and dependencies

Project Status Public Beta

Section titled “Project Status ”

This project is now in public beta. Core functionality is stable and ready for evaluation, though some features are still being refined. Follow progress in the Knowledge Graph First Iteration epic.