What's New

Browse and insert reusable components directly in the Reporting dashboard editor - no need to switch to Development.

Previously, inserting library components required switching to Development, which added friction and blocked Explorers entirely. Now you can insert existing components right where you work - components built by one teammate are instantly available to everyone.

Learn more: Reusable components

Color Palettes are now as-code, meaning you can define and manage your palettes directly within your AML workspace. Now you can:

• Define as code: Create palettes in palette.aml once and reference them anywhere, with full version control.

• Set a workspace-level default: Mark a palette in settings.aml so charts automatically use it by default, unless overridden at the theme or visualization level.

(*) Legacy palettes remain usable and deletable, but they can no longer be edited. To update a legacy palette, recreate it in AML.

Learn more: Color palettes | AML syntax reference

Previously, KPI styling had to be set widget by widget and had limited styling options.

🌟 What's new​

We're bringing KPI metric styling to the dashboard theme and supporting much richer styling options.

• Style once, apply everywhere: Configure in the dashboard theme, and changes will be applied to every KPI across the dashboard. Learn how to style with dashboard theme.

• More styling options: Customize alignment, label and value text, progress bars (text, indicator, track), and trend indicators per state (positive, negative, neutral). Learn how to customize your KPI metric.

Embed Events lets you control embedded Holistics dashboards from your application using JavaScript. Send commands, listen for user interactions, and keep your app and dashboard in sync — all through the browser's standard postMessage API.

Previously, embedded dashboards were one-directional — your app could configure the embed at load time but couldn't respond to user interactions or send commands afterwards. With Embed Events you can:

• Sync filters — push filter values from your app into the dashboard, or listen for user filter changes and reflect them in your app
• Navigate dashboards — build a custom tab switcher or menu that controls the iframe without a full reload
• React to errors — detect embed_error events and auto-refresh an expired JWT to keep the session alive

Learn more: App & Embed Communication

Embedded dashboards can now re-run automatically whenever filters or controls change, without clicking the Apply button. Use the new dashboard_autorun_on_changes setting in your JWT payload to enable this behavior across all dashboards in your embed portal.

How to enable​

Add dashboard_autorun_on_changes: true to the settings object in your JWT payload:

const payload = {
  object_name: 'your_portal',
  object_type: 'EmbedPortal',
  settings: {
    dashboard_autorun_on_changes: true,
  },
  exp: Math.floor(Date.now() / 1000) + 3600,
};

This overrides the dashboard's internal auto-run on changes setting. The default is false — dashboards continue to require clicking Apply unless you opt in.

Note: Auto-run fires a query on every filter change. For dashboards with many charts or heavy queries, leaving this off avoids unexpected query costs.

Learn more: Embed Portal Parameters Reference

Power BI to Holistics migration is now available. Holistics maps every core Power BI concept to its Holistics equivalent, and a new migration skill applies that mapping for you in a single command.

Here's how the concepts line up:

• Semantic models → datasets, with table relationships carried over (including role-playing relationships)
• DAX measures → AQL metrics
• Reports → Canvas Dashboards, filters and visuals included

Previously, switching from Power BI meant rebuilding all of that by hand, which is weeks of work that kept teams from making the move. Now you drop your .pbip export next to a Holistics project and run one command. The skill writes the models, translates the metrics, rebuilds the dashboards, validates parity against your original Power BI values, and syncs to Cloud. Weeks become hours.

Learn more: Migrating to Holistics from Power BI

The Explore interface has a new look and feel as part of our ongoing glow-up.

New chart selector - The chart picker now surfaces the most common chart types with a clearer icon set that makes it easier to distinguish between chart types at a glance. Less common charts are tucked behind a "more" option to keep the interface clean.

Configure button - The viz settings panel now opens and closes via a Configure button in the toolbar area.

Toolbar - The toolbar is where you can perform most Explore actions: ask AI, break down your data, drill by time, and more - all without opening the viz panel.

Run button in the toolbar - The Run button has a new home in the top-right toolbar alongside Analyze and Ask AI. This is the first step toward a workflow where you can control charts entirely from the field list and toolbar, without needing to open the viz panel.

Learn more: Explore is getting a glow-up

Local agentic development setup is now a single skill. After cloning your project, open Claude Code and run /setup-amql-development and it handles the whole setup for you.

Setting up a local development workflow used to mean a string of manual steps, each with its own way to fail:

• Set up Git integration and pull the project locally
• Install the Holistics CLI and authenticate
• Connect the MCP server and authenticate
• Install the Holistics skill
• ...and only then start developing

The new /setup-amql-development skill collapses that into one step. Once you clone your project, install the holistics/skills plugin, run the skill, and wait for it to finish. Under the hood it:

• Installs and authenticates the Holistics CLI over OAuth (no more API key copying)
• Connects the MCP server to your agent
• Starts a continuous holistics sync-code session

When it completes, your agent connects to your workspace and you can start developing.

Learn more: Local development with AI agents

Select all fields in a model is now available in the field picker in Holistics.

Previously, selecting every field in a model meant checking each one individually — tedious for large models and a common friction point when exploring a dataset for the first time. Now you can select or deselect all fields in a model with a single click.

A checkbox appears next to each model name in the field picker. Clicking it selects all fields in that model. Click again to deselect.

The same behavior extends to folders and sub-folders, so you can bulk-select all fields within those groups at once.

The checkbox is available for Table visualization only.

Oracle Database is now a supported data source in Holistics. You can connect your Oracle warehouse directly and build models, datasets, and dashboards with full filter support, no ETL workaround required.

Previously, Oracle-shop teams had to land their data in a separate warehouse before Holistics could query it. Now you can point Holistics straight at Oracle and get to work.

Learn more: Oracle