AI Sandboxing

Build, run and test the AI’s code in a virtual machine, all within your browser.

What is it?

Fine’s AI Sandboxing runs your code in a secure dev environment, allowing you to test the AI’s code practically and visually. You can run the newly generated code and view the console logs and a preview from within your browser, making the AI coding workflow more seamless than ever.

Why AI Sandboxing of AI generated code is so important for developers

Imagine a world in which AI not only takes your specs and writes the code, but can run and test the code, compare it to the specs and fix itself. Where AI not only writes code that looks right and makes sense to the LLM’s logic, but verifies that it actually works - and meets the exact requirements you defined.

AI Sandboxing is the next step in achieving this.

With the help of your feedback on the current version of AI Sandboxing, we’ll soon be setting our sights on releasing a complete self-assessing, self-improving AI coding agent - the first ever.

How to use it

Configuring your AI Sandboxing Settings

In order to be able to run a project, you’ll need to set up the environment.

Head to Projects and click the Settings icon on the Project (repository) you’d like to configure.

You need to provide the AI agent the following information:

• Install script - (e.g. npm install or pip install - requirements.txt) to install project’s dependencies
• Run script (e.g. npm run dev)
• Application port (e.g. 8080) for port-forwarding and external access

Optionally, you can configure environment variables that are needed to run your app. Environment variables are securely stored in a secret management service.

Your application runs within an isolated dev container environment which is preconfigured with popular dev tools (e.g. Python 3.11 and Node 18) that works for most projects.

Note: In the future we will support custom dev environments.

Running the Code

To run the code generated by the AI, simply click the green Run button in the top right corner of your screen. The console will open and automatically draw all the necessary files.

View the Preview

When the code has finished running, click “Open” under the Preview tab to view the result. The preview will open in a new tab in your browser.

You can copy and paste the live preview URL to share with colleagues.

Making further changes

You can continue the conversation with the AI to make further revisions to the code, based on what you’ve seen in the preview.

Describe the issue or change in the conversation box and click enter.

Currently, the logs you see in the console are not included in Fine’s context - that is to say, the AI won’t know what has or hasn’t worked unless you tell it. However, you can copy and paste the logs into the conversation to guide the AI further. (We soon will be releasing another update, in which it will happen automatically in case of an error).

Accessing AI Sandboxing from issues in Linear and GitHub

When you use the /code function in GitHub Issues or Linear, and when you add the Fine label to issues, you can click “Track Progress” in Fine’s reply to open the conversation in your browser, where you’ll be able to use the AI Sandboxing feature.

AI Sandboxing is a Pro feature of Fine

But at just $13-15 per month, it’s a bargain.