Terms and acceptable use

Plain language, deliberately. Last updated 18 July 2026.

You are working with an AI system

Your instructions are sent to a large language model that writes and edits code on your behalf. You are not talking to a person. The model can be wrong, confidently — it can misunderstand a request, introduce a defect, or make a choice you did not intend. Review what it produces before you rely on it, and before you put real user data into anything it built.

We run automated typecheck, lint, and test gates on generated code and will not mark a change complete until they pass. That catches broken code. It does not catch code that works and is still wrong for your purpose.

What you may not build

These rules apply to you and to anything you build here:

• Phishing, credential harvesting, or impersonating a real person or business.
• Malware, or infrastructure for attacking others.
• Spam, bulk unsolicited messaging, or evading a platform's controls.
• Content sexualising minors, or unlawful content of any kind.
• Mining cryptocurrency, or using build capacity for unrelated computation.
• Infringing someone else's copyright or trademark.

Applications live on a shared domain. One abusive site can get every customer's URL blocklisted, which is why we suspend clear cases immediately rather than after a conversation.

Your code and your data

Your project's code is yours. You can clone it and take it elsewhere at any time. We keep the version history so you can go back to an earlier state.

Each project gets its own database with its own credentials, isolated from every other project. We can read your project's data only to operate the service or to investigate an abuse report.

Credits

Every workspace starts with free credits. Building consumes them based on the AI usage a turn required — you can see the exact figure for every turn. If our own automated checks reject our output and we retry, that repair is not charged to you.

Availability

This is early software with no uptime guarantee. Preview URLs are for development and can change. Do not put anything irreplaceable here without your own backup.

Reporting problems

Abuse: report it here. Security: security@startnicely.com.

This document is written to be readable rather than exhaustive, and has not yet been reviewed by a lawyer.