# Acceptable Use Policy

> What you may not do with Roasby, so that the service stays safe for you, the ad platforms and everyone else.

This Acceptable Use Policy is part of the [Terms of Service](/terms) between you and Roasby. It applies to everyone who uses Roasby, including the AI assistants and automations acting for you. "You" includes your workspace members and the assistants you connect. The rules below are examples of what is not allowed, not a complete list: we may also act on other uses that harm Roasby, its users, the ad platforms or others.

## Advertising

Do not use Roasby to create, edit or run ads that:

- break the law, or the advertising policies of the platform they run on (Google Ads, Microsoft Advertising, ChatGPT Ads or others);
- are deceptive, fraudulent or misleading, including fake endorsements, scams, phishing or impersonation;
- promote illegal products or services, or target people in ways the law or the platform forbids;
- infringe anyone's intellectual property, privacy or publicity rights;
- contain malware or link to it.

Do not use Roasby to get around a platform's enforcement: for example, to manage an account the platform suspended, to create accounts to evade a ban, or to hide what an ad really promotes (cloaking).

## Accounts and access

- Connect only ad accounts you own or are authorized to manage, and stop managing a client's account when your authority ends.
- Do not share sign-in details, tokens or invitations with people who are not workspace members.
- Do not access, or try to access, another workspace's data or accounts.

## Automation and AI assistants

Roasby is built for assistants, within limits that protect your money and the platforms:

- Do not try to bypass confirmations, budget guardrails, rate limits, plan limits or other safety features, or to make an assistant confirm changes without a person's review where Roasby asks for one.
- Do not overload the Service or the platforms' APIs, for example with automated loops or excessive refreshes.
- Do not scrape the Service, or use it to build a competing product or dataset.
- Do not use Roasby, or data from it, to train AI models in ways that break a platform's terms (for example, Google's Limited Use requirements).

## Security and integrity

- Do not probe, scan or test the security of Roasby without our permission. To report a vulnerability, follow our [Security page](/security); good-faith research under it is welcome.
- Do not reverse engineer the Service, except where the law allows it despite this restriction.
- Do not upload malicious code, or interfere with the Service, its users or its providers.

## Resale

Do not resell, sublicense or white-label Roasby without a written agreement with us. Agencies may use Roasby to manage their clients' accounts as part of their own services.

## Enforcement

If we believe this policy has been broken, we may remove content, pause automation, limit or suspend access, or end the account, as set out in the Terms. Where an ad platform requires it, we may report abuse to that platform. Where possible, we will tell you first and give you a chance to fix the problem.

## Reporting abuse

Report abuse to **support@roasby.com** with "Abuse" in the subject (or with the [contact form](/contact)), and security issues to **security@roasby.com**.