Terms of Service

1. Acceptance

By using EvalKit you agree to these terms. If you do not agree, do not use the service.

2. What EvalKit provides

EvalKit is a registry and marketplace for AI evaluation packages. We provide infrastructure for publishing, discovering, and running eval suites. We do not execute model calls on your behalf — all inference uses your own API keys (BYOK).

3. Your responsibilities

You are responsible for:

• All content you publish to the registry
• Your own API keys and associated costs
• Ensuring your eval packages do not violate any third-party rights
• Keeping your account credentials secure

4. Prohibited use

You may not use EvalKit to:

• Publish malicious, deceptive, or harmful packages
• Abuse the registry with automated bulk publishing
• Attempt to reverse-engineer or compromise the platform

5. Packages and content

Published packages remain owned by their authors. By publishing, you grant EvalKit a license to host and distribute your package. We reserve the right to remove packages that violate these terms.

6. Service availability

EvalKit is provided as-is during early access. We do not guarantee uptime or data retention at this stage. We will provide reasonable notice before any major service changes.

7. Limitation of liability

EvalKit is not liable for any indirect, incidental, or consequential damages arising from your use of the service, including costs incurred through your own LLM API keys.

8. Changes to these terms

We may update these terms. Continued use after changes constitutes acceptance. We will notify users of material changes via email or dashboard notice.

9. Contact

Questions about these terms: contact email coming soon.