Local development with external IDEs
Introduction
Holistics is built for development outside the UI. Everything in Holistics — models, datasets, metrics, and dashboards — is defined as code, version controlled in Git, and can be managed through CLI and API.
This guide walks you through a typical local development workflow using your preferred IDE.
Prerequisites
Before you begin, make sure you have:
- A Holistics project connected to an external Git repository (GitHub, GitLab, Bitbucket, etc.)
- Git installed on your local machine
- An IDE of your choice (VS Code, Cursor, etc.)
Development workflow
Here's the typical workflow for developing Holistics analytics code locally:
1. Clone your repository
Clone your Holistics project repository to your local machine:
git clone [email protected]:your-org/your-holistics-repo.git
cd your-holistics-repo
2. Create a development branch
Create a new branch for your changes:
git checkout -b feature/add-new-dashboard
3. Make changes in your IDE
Open the project in your preferred IDE and start editing AML files. Your project structure typically looks like:
your-holistics-repo/
├── models/
│ ├── orders.model.aml
│ └── users.model.aml
├── datasets/
│ └── ecommerce.dataset.aml
└── dashboards/
└── sales_overview.page.aml
4. Push and preview
Push your changes to your branch:
git add .
git commit -m "Add new sales dashboard"
git push -u origin feature/add-new-dashboard
In the Holistics web UI, switch to your branch to preview your changes before merging.
5. Validate your code
Validate your changes using one of these methods:
-
CLI validation for local validation:
holistics aml validate -
Validation API for project-level validation in CI/CD pipelines
Both integrate cleanly into CI/CD workflows — for example, a GitHub PR workflow that validates on every commit.
6. Create a pull request
Create a pull request either:
- Locally via your Git tooling (
gh pr create, GitLab CLI, etc.) - Through the Holistics web UI
7. Code review and merge
Have your team review the changes. Once approved, merge to your main branch.
8. Deploy
Deploy using the Holistics web UI, or set up a GitHub workflow to auto-publish via the Publish API.
IDE tooling
VS Code extension
We offer a VS Code extension that enhances your local development experience with:
- Syntax highlighting for AML files
- Autocomplete for AML keywords and properties
- Hover information showing documentation for AML elements
- Go-to-definition to navigate between models, datasets, and dashboards
To install, search for "Holistics" in the VS Code Extensions marketplace, or run:
code --install-extension holistics.holistics-aml-vscode-ext
Other IDEs
AML files are plain text, so any IDE works for editing. Popular choices include:
- Cursor — AI-powered IDE with good support for custom languages
- JetBrains IDEs — IntelliJ, WebStorm, etc.
- Vim/Neovim — For terminal-based workflows
For AI-assisted development with tools like Claude Code or Cursor, see MCP Server for AI Agents.