klovys99 sits between your employees and LLMs (ChatGPT, Claude, Gemini) to anonymize sensitive data, monitor interactions, and cache tokens. A sovereign European cloud solution, compliant with GDPR and the AI Act.

👋 Hi Product Hunt! We're three backend developers who noticed the same thing happening everywhere: AI adoption is exploding, but companies have almost no visibility into what employees are sending to ChatGPT, Claude, Gemini, and other LLMs. Most organizations don't want to ban AI, they just want to use it safely. That's why we built klovys99, an open-source proxy that sits between users and LLMs to: * 🔒 Anonymize sensitive data before it leaves the company * 👀 Give organizations more visibility and governance over AI usage Our goal isn't to replace GPT or Claude with a self-hosted model. We believe companies should be able to use the best LLMs available while keeping control of their sensitive data and staying compliant with regulations like GDPR and the AI Act. We're still at a very early stage, so we'd genuinely love your feedback, whether it's about the architecture, security model, feature ideas, or anything else. Every comment helps us build a better product. Thanks for checking us out! ❤️

the token caching actually cut our api costs noticeably after a week of testing, and the gdpr compliance angle is a real plus for our eu clients

Congrats on the launch. One thing I'd love to see is a Slack or Teams notification whenever the anonymizer flags a request that could leak PII, so admins get real-time alerts instead of only catching it in the dashboard later.

the token caching actually cut our api costs noticeably after a week of testing, and the gdpr compliance angle is a real plus for our eu clients

klovys99

klovys99 is a proxy to anonymize sensitive data

klovys99 is a proxy to anonymize sensitive data