Project Templates

Vaibify ships with two project templates that provide starting configurations for common use cases. Select a template when initializing a new project:

vaibify init --template <name>

Each template creates three files in the current directory: vaibify.yml, container.conf, and workflow.json.

sandbox

An empty starting point for exploration, prototyping, and interactive use. No pipeline steps are defined — you work directly inside the container.

Includes:

Minimal vaibify.yml with default Python version and base image.
Empty container.conf (no repositories).
Empty workflow.json (no pipeline steps).

Use this template when you want a containerized environment without a predefined workflow.

workflow

A starting point for reproducible data analysis pipelines. Includes an example step with data generation and plotting commands that you replace with your own.

Includes:

Minimal vaibify.yml with default Python version and base image.
Empty container.conf (add your repositories).
Example workflow.json with one step (AnalyzeData) that runs python runAnalysis.py and python makePlot.py.

Use this template when your project follows a defined sequence of analysis steps that should be reproducible.

Adding LaTeX compilation: If you compile your manuscript inside the container rather than using an external tool like Overleaf, add a step to workflow.json:

{
    "sName": "CompileManuscript",
    "sDirectory": "tex",
    "bEnabled": true,
    "bPlotOnly": false,
    "saDataCommands": [],
    "saDataFiles": [],
    "saPlotCommands": ["latexmk -pdf manuscript.tex"],
    "saPlotFiles": []
}

Creating Custom Templates

Templates are stored in the templates/ directory of the Vaibify package. Each template is a subdirectory containing vaibify.yml, container.conf, and workflow.json. To create a custom template, add a new subdirectory with these three files and reinstall the package.