Hey Product Hunt 👋 I’m the maker of Xytha. I started building Xytha around a simple question: As AI agents begin working across different tools and platforms, why should their identity and work history reset every time they move somewhere new? Xytha is a portable trust layer for humans, AI agents, and teams. Today, it supports stable agent identities, version-specific credentials, signed collaboration evidence, transparency proofs, and evidence-backed trust profiles. Humans and AI agents can also form teams and turn completed work into verifiable collaboration records. One principle is especially important to me: Xytha provides evidence, not verdicts. We don’t want to assign a universal score that decides whether a person or agent is “good.” Instead, Xytha helps show what happened, which agent version performed the work, who participated, and what evidence exists. The Public Preview is now live, including account registration and a Windows Trust Workbench. I’d really value feedback on three things: 1. Would portable work history for AI agents be useful to you? 2. What evidence would you want before trusting an unfamiliar AI agent? 3. Which parts of Xytha feel useful, confusing, or unnecessary? Thanks for checking it out. I’ll be here reading and responding to feedback.

Xytha

Portable trust for humans, AI agents, and teams

Portable trust for humans, AI agents, and teams