No installation required • Instant visualization • Works in your browser
Analyze and visualize your codebase dependencies directly in your browser.
Upload your .cg.json file or explore with sample data.
Fully client-side • Your data never leaves your machine • Privacy guaranteed
Having trouble understanding the structure and dependencies in your codebase? DependaCharta is a multi-language code analysis and visualization tool that helps you quickly identify disadvantageous dependencies, cyclic dependencies, and quality hotspots. It runs entirely locally, works with non-compilable code, and supports multiple programming languages. DependaCharta is an open source project developed by MaibornWolff.
For a detailed explanation of DependaCharta's domain concepts and visualization features, see our Domain Model Documentation.
All analysis and visualization happens entirely on your machine. No data is ever uploaded, shared, or transmitted to any external service. Your code stays private and secure. DependaCharta has no analytics, tracking, or telemetry built-in.
Download the latest JAR from the releases page and run:
java -jar dependacharta.jar -d <directory-to-analyze>Or use the convenience scripts included in the release: bin/dependacharta.sh (Mac/Linux) or bin/dependacharta.bat (Windows).
With Docker (no Java required):
docker run --rm --user root \
-v "$(pwd)/your-project:/workspace" \
maibornwolff/dependacharta-analysis:latest \
-d /workspaceThe analysis outputs a .cg.json file (default: output/analysis.cg.json).
- Web Studio — Open your
.cg.jsonfile directly in the browser. No installation required, fully client-side. - Standalone App — Download the desktop app for your OS from the releases page.
- Run locally — Build and run the visualization yourself. See visualization/README.md.
The Analysis component is a command-line tool that analyzes codebases using Tree-sitter parsers. It creates .cg.json files containing structural information about your project, including:
- File and package hierarchies
- Dependencies between files and packages
- Cycle detection with detailed cycle information
- Levelization for architectural insights
- Support for multiple programming languages: Java, C#, C++, TypeScript, JavaScript, PHP, Go, Python, Kotlin, Vue
Don't see your language? You can extend the supported languages by adding a new parser!
The Visualization component is an Angular/TypeScript web application using Cytoscape.js for interactive dependency graphs. Features include:
- Interactive graph visualization of your codebase structure
- Navigate through package and file dependencies
- Filter and highlight critical dependencies
- Identify cyclic dependencies and architecture violations
- Find quality hotspots requiring attention
- Multiple layout algorithms for optimal viewing
- Available as both web app and Electron desktop application
Mise manages both runtime versions (Java, Node.js) and project tasks. One tool, zero manual setup.
# Install mise (https://mise.jdx.dev/getting-started.html)
curl https://mise.run | sh # or: brew install mise
# Trust and install project runtimes (first time only)
mise trust
mise install
# See all available tasks
mise tasks
# Key workflows
mise run analyze <directory> # Analyze a codebase
mise run analyze-and-serve <directory> # Analyze and open visualization
mise run test-analysis # Run analysis unit tests
mise run dev-visualization # Start development serverTip: Run mise tasks to discover all available commands.
If you prefer not to use mise, install the prerequisites manually:
- Java: 17 or higher (
java -version) - Node.js: 22 or higher (
node -v)
cd analysis
# Build the analysis tool
./gradlew fatJar
# Run the analyzer
java -jar build/libs/dependacharta.jar -d <directory-to-analyze>
# Or use the convenience scripts:
# bin/dependacharta.sh (Mac/Linux)
# bin/dependacharta.bat (Windows)
# Run tests
./gradlew testFor detailed analysis documentation, see analysis/README.md.
cd visualization
# Install dependencies
npm ci
# Development server (http://localhost:4200)
npm run start
# Run as Electron app
npm run start-electron
# Build for production
npm run build
# Run tests
npm run test
# Package for distribution
npm run package-win # Windows
npm run package-mac-silicon # macOS ARM64
npm run package-mac-intel # macOS x64For detailed visualization documentation, see visualization/README.md.
DependaCharta provides official Docker images via Docker Hub for easy deployment and CI/CD integration.
# Pull the latest image
docker pull maibornwolff/dependacharta-analysis:latest
# Run analysis on a directory
docker run --rm --user root \
-v "$(pwd)/your-project:/workspace" \
maibornwolff/dependacharta-analysis:latest \
-d /workspace
# Output will be in your-project/output/analysis.cg.json# Build Docker image
mise run docker-build
# Run analysis with Docker
mise run docker-run <directory-to-analyze>
# Analyze and view results
mise run docker-analyze <directory-to-analyze>Note: When using Docker on macOS/Windows, run the container with --user root to avoid permission issues with mounted volumes.
Do you have a bug, feature request, or question? Please open a new issue. Feedback is always welcome.
Want to contribute? Check out our contributing guide and developer start guide.
- Test-driven development with tests before implementation
- Clean Code principles (pragmatic, not dogmatic)
- Semantic Versioning for releases
- Automated CI/CD pipeline (see PIPELINE.md)
- Releases
- Domain Model Documentation
- Analysis Documentation
- Visualization Documentation
- How to Add a New Language
- Cycle Detection Algorithm
- Levelization Algorithm
- Architecture Decision Records
- Pipeline Documentation
BSD-3-Clause License
Made with ❤ by MaibornWolff · GitHub @MaibornWolff