I built Viatoris (https://viatoris.ai) because I kept running into the same problem in enterprise tech: companies deploy AI agents, then can't prove what those agents did when compliance or a customer asks. Logs aren't enough. They're mutable and unverifiable. Viatoris gives every agent a cryptographic identity (W3C DID + Ed25519 key pair). Every action produces a signed receipt. Tamper with it and the signature breaks. Anyone can verify an agent's identity and history through a public API. Integration is one line if you're using LangChain, CrewAI, or AutoGen. We also have TypeScript and Python SDKs for custom setups. It's live, free tier available. I'm a solo founder building this from enterprise tech experience. Would love feedback on whether this solves a real problem for anyone here, and what I'm missing. https://viatoris.ai https://viatoris.ai/quickstart
Comments URL: https://news.ycombinator.com/item?id=47698941
Points: 2
# Comments: 0