How to Set Up Rill in Your Desktop Step by Step

Running analytics dashboards locally gives you development speed. You can test changes, pull credentials, and see results without waiting for cloud deployments.

We recently walked through the complete process of setting up Rill Developer on a desktop machine. This workflow follows what most developers use when they need to run Rill dashboards locally for development and testing purposes.

Here's the complete technical walkthrough of how to set up Rill in your desktop.

Key takeaways

• Installing Rill Developer takes one command in your terminal and gives you the Command Line Interface (CLI) tools you need
• Cloning the Rill repo from GitHub Desktop or terminal brings your dashboard project to your local machine
• Pulling environment credentials from Rill Cloud provides the Snowflake connection details your dashboard needs
• Running rill start launches your dashboard locally, but only after you've pulled credentials first
• Forgetting to pull credentials before running rill start means your dashboard loads without any data

Problems solved by local setup

Before running Rill locally, developers face challenges that slow down their workflow and create friction in the development process.

Dashboard changes require cloud deployments

Testing even small changes means pushing to Rill Cloud and waiting for deployments. Each iteration takes time. You can't rapidly test different configurations or debug issues when every change requires a full deployment cycle.

Environment setup isn't clear

The connection between Rill Cloud credentials and your local environment creates confusion. Developers wonder which credentials they need, where those credentials live, and how to get them onto their local machine without manual copying.

Missing data creates debugging confusion

Running rill start without proper credentials gives you an empty dashboard. The dashboard loads successfully but shows no data. This creates uncertainty about whether the issue is credentials, data sources, or something else in your configuration.

These challenges mean developers spend more time on deployment cycles than on actual development work.

Install Rill Developer first

You need Rill Developer installed before you can run anything locally.

The installation uses a single curl command that downloads and installs everything. This command works on Mac and Linux systems. You run this command in your terminal, whether that's the built-in terminal or one inside Cursor or VS Code.

After installation completes, you check the versions to confirm Rill installed correctly. This verification step ensures the CLI tools are available and working.

The installation process takes about 30 seconds. Once complete, you have access to all Rill commands from your terminal.

Clone the repo from GitHub

Every Rill dashboard project lives in a repository. You need that repository on your local machine.

Using GitHub Desktop

GitHub Desktop provides a visual interface for cloning. You find your Rill dashboard repository in GitHub, then click "Open with GitHub Desktop." The application handles the clone process and places the folder in your designated GitHub directory.

After cloning through GitHub Desktop, you open Cursor or VS Code and select "Open Folder." You navigate to where GitHub Desktop stored the repository and open that folder.

Using terminal commands

You can also clone using git commands directly in the terminal. Copy the repository link from GitHub, then use standard git clone commands. This approach gives you more control over where the repository goes.

Either method works. Choose the approach that matches your workflow.

Open the project in your editor

Cursor and VS Code provide the same functionality for this workflow. Both editors give you an integrated terminal where you can run Rill commands.

You open the folder containing your cloned repository. The folder structure shows your dashboard files, configuration files, and the project setup.

The project name in GitHub matches the folder name on your local machine. This naming consistency matters when you run commands that reference the project.

Pull environment credentials

This step connects your local setup to Rill Cloud credentials. Without these credentials, your dashboard can't access Snowflake tables or other data sources.

Understanding the env file

The environment file contains all the credentials your dashboard needs. These credentials define connections to Snowflake, authentication details, and other configuration that shouldn't live in your repository.

Rill Cloud stores these credentials securely. The pull command downloads them to your local machine as an env file.

Running rill env pull

You run the command rill env pull followed by your project name. The project name matches your repository name and folder name.

This command connects to Rill Cloud, authenticates your access, and downloads the environment file with all your Snowflake credentials and data source configurations.

You see confirmation in the terminal when the pull succeeds. The env file now exists in your project folder, though it won't be visible in most file explorers because it's a hidden file.

Why credentials matter

Without the env file, rill start launches a dashboard that shows the interface but no data. The dashboard code runs successfully, but data queries fail silently because they can't connect to Snowflake tables.

You must pull the environment credentials before you run rill start. If you forget this order, you need to kill the project and start over after pulling credentials.

Run rill start command

This simple command launches your dashboard locally. You type rill start in your terminal and press enter.

The command starts the Rill server and opens your dashboard in a browser. You see the startup process in the terminal as the server initializes.

What happens during startup

Rill reads your project configuration, connects to data sources using the credentials from your env file, and starts serving the dashboard interface. The process takes a few seconds.

Your browser opens automatically to localhost with a port number. The dashboard loads and displays your data if you pulled credentials correctly.

Viewing your dashboard

Once rill start completes, you see your dashboard with all its components. Multiple dashboards within one project appear as separate views. You can navigate between them using the dashboard interface.

The dashboard shows real data from your Snowflake tables because the env file provided the necessary connection credentials. Charts, tables, and metrics all populate with current data.

Troubleshooting missing data

The most common issue is running rill start before pulling environment credentials.

Dashboard loads but shows no data

This symptom indicates the dashboard code works but data connections failed. The interface renders correctly. Navigation functions properly. But every chart and table remains empty.

You probably ran rill start before running rill env pull. The dashboard launched without the credentials needed to connect to Snowflake.

How to fix the issue

Kill the running project in your terminal. Use Control+C to stop the rill start process.

Then run rill env pull [project-name] to download the environment credentials. Wait for confirmation that the pull succeeded.

Now run rill start again. The dashboard will load with data this time because it has the credentials it needs.

Remember the order

Always follow this sequence: install Rill Developer, clone the repo, pull the environment credentials, then run rill start. Pulling credentials before starting prevents the empty dashboard problem.

Complete setup workflow

This end-to-end process represents the standard approach for local Rill development:

Installation: Run the curl command to install Rill Developer and check versions

Repository: Clone your Rill dashboard repo using GitHub Desktop or git commands

Editor: Open the project folder in Cursor or VS Code

Credentials: Pull environment credentials from Rill Cloud with rill env pull

Launch: Start the local dashboard with rill start

Verification: Confirm data displays correctly in the dashboard interface

The workflow takes about five minutes after you know the steps. Each step depends on the previous one completing successfully.

FAQ

What if I don't have GitHub Desktop installed?

You can clone repositories using git commands directly in the terminal. GitHub Desktop provides a visual interface but isn't required. Standard git clone commands work the same way.

Can I use VS Code instead of Cursor?

Yes. Both editors provide integrated terminals and identical functionality for this workflow. Cursor and VS Code handle Rill commands the same way.

Why does my dashboard show no data even after rill start?

You probably ran rill start before pulling environment credentials. Kill the project, run rill env pull with your project name, then run rill start again.

Where does the env file go after I pull it?

The env file saves in your project folder as a hidden file. It contains your Snowflake credentials and other sensitive configuration that shouldn't be committed to your repository.

Do I need to pull credentials every time I start the project?

No. Once you pull credentials, the env file stays in your project folder. You only need to pull again if credentials change in Rill Cloud or if you delete the env file.

Summary

Setting up Rill in your desktop enables rapid local development without cloud deployment delays. The five-step workflow covers installation, repository cloning, credential pulling, and dashboard launching.

Environment credentials connect your local setup to data sources. Pulling these credentials before running rill start prevents the common issue of dashboards that load without data.

The combination of Rill Developer, GitHub Desktop, your preferred editor, and the rill start command creates a local development environment that mirrors your cloud deployment while giving you faster iteration cycles.

This approach works for any Rill dashboard project because it follows the standard pattern for moving from cloud deployment to local development in a reproducible way.