Hi Product Hunt! I’m DongGyun, the founder of Styxis and the maker of Truthound Depot. I started building Truthound Depot after seeing data changes move faster than teams could reliably review them. This becomes especially risky with RAG datasets, evaluation data, automated workflows, and AI agents: a small, unnoticed change can quietly affect production behavior. Most tools can store data or show a validation result, but teams still need to understand what changed, decide whether it is safe, and preserve the reasoning behind that decision. Truthound Depot brings that workflow into one place. You can import datasets as encrypted, immutable Data Versions or validate a selected cloud source directly. From there, you can compare changes, run validation, request a review, preserve evidence, publish a verified version, and restore an earlier approved version if needed. Our guiding idea is simple: “Sniff before you merge.” I’d love to hear how your team currently reviews dataset changes, especially for RAG, eval, and production data. Your feedback will help shape what we build next. — DongGyun Founder, Styxis https://styxis-io.com

Truthound Depot Cloud

Validate dataset changes before production

Validate dataset changes before production