Acceptable Use Policy

1. Prohibited content

You may not upload, store, or transmit through the Service:

• content that is illegal under applicable law, including content that infringes intellectual property, trade secrets, or privacy rights;
• malware, ransomware, or any code intended to disrupt or gain unauthorised access to a system;
• personal data of third parties for which you do not have a lawful basis to process;
• content depicting or facilitating the sexual exploitation of children, terrorism, or incitement to violence;
• regulated data — payment card data, government identifiers, protected health information — except where a written addendum has been signed permitting it.

2. Prohibited activity

• scraping the Service, except through the documented APIs and within published rate limits;
• reverse-engineering, decompiling, or attempting to extract source code or models;
• probing, scanning, or testing the vulnerability of the Service except in line with our responsible-disclosure programme;
• interfering with another customer's use of the Service, including attempts to access another workspace;
• circumventing rate limits, authentication, or other technical protections;
• using the Service to send unsolicited bulk email or SMS, or to impersonate another person or organisation;
• reselling or sublicensing the Service without our written consent.

3. AI-specific rules

Shelby is a tool. It is not a person and it is not infallible. You agree:

• not to craft prompts intended to extract content from another tenant, exfiltrate system prompts, or bypass safety controls;
• not to use Shelby outputs as the sole basis for safety-critical, life-critical, or regulatory decisions — outputs must be reviewed by a qualified human before action;
• not to upload third-party copyrighted material for the purpose of training other AI systems;
• that automated, high-volume prompt traffic from non-human agents is prohibited unless agreed in writing.

4. Account and credential hygiene

• do not share credentials between users — each individual using the Service must have their own account;
• notify us promptly at [email protected] if you suspect a credential has been compromised;
• use a password manager. Reusing passwords across services is the single most common cause of breaches.

5. Enforcement

We may investigate suspected AUP violations. Depending on severity, we may (a) ask you to stop, (b) temporarily suspend access, (c) terminate your workspace, or (d) cooperate with law enforcement. Where lawful and operationally safe, we will give notice and a reasonable opportunity to cure before terminating.

6. Reporting

To report a violation of this AUP, email [email protected]. For security issues, use [email protected].

7. Changes

We may update this policy as the product evolves. Material changes are communicated to workspace owners by email at least thirty (30) days before they take effect.