Acceptable use policy
The short version
- Use Managed MCPs for lawful work.
- Do not use it to harm, deceive, spam or attack anyone, including us.
- If you are not sure, ask us first.
Do not use Managed MCPs to:
- break the law or help someone else break it;
- harass, threaten, defraud or impersonate anyone, or pretend that an AI is a human when someone asks;
- send spam or unsolicited bulk messages, or messages to people who have not agreed to receive them;
- collect or process personal data without a lawful basis, or process special-category data without the safeguards the law requires;
- create content that sexualises minors, promotes violence or terrorism, or incites hatred;
- make automated decisions with legal or similarly significant effects on people without proper human review;
- build weapons, malware or tools to break into systems you are not allowed to access;
- probe, scan, overload or get around the security or limits of Managed MCPs or connected services (security research only with our written permission);
- infringe anyone's intellectual property or privacy;
- resell or share access in a way your agreement does not allow.
Model provider rules
Where Managed MCPs uses AI model providers, their usage policies also apply to content you send. We pass on only what is needed to meet them.
What we do about misuse
We may remove content, pause automations, revoke keys or suspend accounts to stop misuse. Where it is safe, we tell you first. Report misuse to [email protected].