Terms of use
MCP Trust Index is operated by Klars AI. By using the site or the public API you accept these terms.
Last updated 30 August 2026.
1. What we publish
We publish an automated index of Model Context Protocol servers built from public sources. Scores measure maintenance activity. Risk flags are heuristic observations drawn from published metadata; nothing is downloaded, installed, executed or manually reviewed. Both are generated by software and may be incomplete, delayed or wrong.
2. No warranty
The service is provided “as is” and “as available”, without warranty of any kind, express or implied, including fitness for a particular purpose, accuracy and non-infringement. We do not guarantee that the index is complete, current or uninterrupted.
3. Limitation of liability
To the fullest extent permitted by law, Klars AI is not liable for any indirect, incidental, special or consequential loss, or for any loss of profit, data or business, arising from use of or reliance on this service. Nothing in these terms excludes liability that cannot lawfully be excluded — including for death or personal injury caused by negligence, or for fraud.
4. Using the data
Index data served through this site and the public API is licensed under CC BY 4.0. You may use, adapt and redistribute it, including commercially, provided you credit “MCP Trust Index” with a link back. The site design, code and written content are separately copyright Klars AI.
5. Fair use of the API
The public API is free and unauthenticated. Please cache responses, identify your client with a real user agent, and keep to a reasonable request rate. We may rate-limit or block clients that degrade the service for others, and may introduce authentication and quotas in future — existing endpoints will remain versioned under /api/v1.
6. Publishers and corrections
Listing is automatic, drawn from the public registry, and is neither an endorsement nor a criticism. If your server is listed and a record is inaccurate, email contact@klars.ai and we will correct it. We will consider reasonable removal requests from verified publishers.
7. Acceptable use
Do not attempt to disrupt the service, circumvent rate limits, scrape in a way that degrades it for others, or present the data as your own assessment or as a security certification.
8. Trademarks
Model Context Protocol, Anthropic, GitHub and all server and product names are the trademarks of their respective owners. MCP Trust Index is an independent project and is not affiliated with, endorsed by or sponsored by any of them.
9. Changes and governing law
We may update these terms; the date above will change with them, and continued use constitutes acceptance. These terms are governed by the laws of England and Wales.
10. Contact
Questions about these terms: contact@klars.ai.