Navigating Agentic Standards: What to Build On Now and What to Watch

Teams putting agents into production keep meeting the same pile of names: WebBotAuth, OAuth, WIF, WAG, ID-JAG, MCP, A2A, AAuth, PACT/MoLE, and at least seven payment protocols. They are being worked out in different venues by different communities, and few teams have time to find out which ones matter or how they fit together.

This talk sorts them by the question each one answers. Which agent is this? On whose behalf is it acting? How does it reach tools and other agents? Should a site trust its traffic? For each layer we'll talk about where the work is happening and how far along it is: what has shipped and is stable enough to depend on, where separate efforts are converging, and what is still open.

Key Takeaways:

1. A framework for thinking about how to make your AI systems interact well with the rest of the world.
2. The agentic standards landscape is complicated, but we've already learned a lot about what works well and what is likely to lead projects astray.
3. In many ways, the rise of agentic AI has made existing best practices more important than ever, rather than creating new problems out of whole cloth.
4. Many of these discussions are happening in the open, and you can follow along or participate.