Acceptable Use Policy

This Acceptable Use Policy applies to everyone who uses Rustbox, including the playground, API, SDKs, dashboard, and any code or input submitted for execution. It supplements the Terms and Conditions.

Rustbox exists to execute untrusted code inside kernel-isolated sandboxes. Running untrusted code inside the sandbox is the intended use. Using the service to attack the platform, its infrastructure, or third parties is not.

You may submit code, including untrusted or experimental code, to run within the documented sandbox limits for development, evaluation, automation, competitive programming, education, agent tooling, and similar legitimate workloads.

You may test the behavior of your own submitted workloads, provided the testing stays within documented limits and does not attempt to escape the sandbox, attack the host, attack the service, or target third parties.

You may not use Rustbox to distribute malware, ransomware, or other malicious code intended to harm third parties, or to host command-and-control infrastructure.

You may not mine cryptocurrency, run sustained proof-of-work, or use the service primarily to extract free compute in a way that is inconsistent with the plan you purchased.

You may not conduct denial-of-service or denial-of-wallet activity, scan, probe, or attack third-party systems, exfiltrate data you are not authorized to access, or send spam, phishing, or unsolicited bulk messages.

You may not attempt to escape the sandbox, compromise the host, attack Rustbox infrastructure, evade authentication or rate limits, abuse signup or quota systems, or interfere with other users' workloads.

You may not submit content that is illegal, infringing, or that you are not authorized to process, and you may not use the service in violation of applicable law or third-party rights.

Sandboxes reach the internet only through the Rustbox egress proxy, on port 443. Each sandbox has a policy: sealed (no network), essentials (package registries, git hosts and model APIs), an allow list of hostnames you choose, or open. The Free plan gets essentials; paid plans get open by default. You can narrow any sandbox's policy at any time.

Every connection is checked against the policy before the name is looked up, and traffic is metered against your plan's limits.

Orkait may block, rate-limit, throttle, suspend, delete, or investigate submissions, API keys, sessions, projects, or accounts that appear abusive, unsafe, fraudulent, or operationally risky, with or without notice where the risk requires it.

The egress proxy suspends a sandbox's network automatically when, within one minute, it connects to more than 150 different hosts or retries more than 300 connections its policy refuses. Those patterns look like scanning and probing, not builds or agents. The sandbox is sealed, and the API key can only create sealed sandboxes until we review it. You can see every suspension, with the numbers that triggered it, at GET /api/egress/suspensions.

If you think a suspension is a mistake, write to [email protected] with the API key's id and what the workload does; a person reviews every one. Serious or repeated violations may result in termination of access and, where appropriate, referral to law enforcement.

If you believe Rustbox is being used in violation of this policy, report it to [email protected] with enough detail for us to investigate.

We may update this Acceptable Use Policy as the service, threat landscape, or legal requirements change. The last-updated date identifies the current version, and continued use after an update means the updated policy applies.