windley.com

Summary Version 1.5 of the Pico Engine finally brings identity into the engine itself: passkeys for humans, and OAuth for external apps and webhooks. Here's the shape of the design, why identity is really three problems and not one, and why I shipped the human and third-party layers before the pico-to-pico layer. Today I'm releasing version 1.5 of the Pico Engine.

Summary In agentic systems, the principal, action, and resource are often unknown until the moment an agent acts, and the context that governs the decision arrives as a flood of signals from many sources. This post looks at how that context takes shape, where each signal is actually consumed, and why a non-directed world of agents still needs decisions that humans can inspect and predict. This post is part of a series on using dynamic authorization to control and coordinate AI agents.

Summary Cleaning up manifold-api as a prerequisite for the spring conversational interface capstone turned into a complete platform update: Pico Engine 1.0 compatibility, automated bootstrap, centralized notifications, and a Docker-based integration test harness. Once the platform was solid, the old temperature-network had an obvious new home inside Manifold's community framework, so I rewrote it too as an example of how Manifold can be a framework for pico networks.

Summary Picos already have persistent identity, owned state, and an event-driven architecture—exactly the properties that make a good substrate for AI agents. The integration path starts with a simple webhook and leads somewhere much more interesting: a world where AI works for you, reasoning over data that is stored in your picos rather than on someone else’s platform.

Summary IIW XLII brought 287 people to the Computer History Museum in Mountain View for three days of sessions on identity, agents, and the legal and technical foundations of first person digital life. The agenda reflected a community grappling with real deployment challenges: SEDI and duty of loyalty, agentic identity, MyTerms, post-quantum cryptography, and the EUDI wallet. AIW2 followed on Friday, continuing the agentic internet conversation.

Summary Alan Mayo frames the digital identity design choice as enhance, duplicate, or replace, and places Utah's SEDI in the "replace" bucket alongside purist decentralized identity. That badly misreads the architecture and the policy goal. SEDI is not trying to eliminate institutional trust; it is state-endorsed, rights-first digital identity reuse that keeps institutional authority where it belongs while moving presentation and consent closer to the individual.

Summary SEDI's duty of loyalty provision shifts the basis for regulating online interaction from the data to the relationship. Where GDPR and similar frameworks treat personal data as the object to be governed, duty of loyalty treats the relationship between the individual and the organization as the thing that matters. MyTerms gives that relationship concrete, operational rails. I'm sitting in a session at IIW hosted by Sam Smith on the duty of loyalty.

Summary MyTerms, the new IEEE 7012 standard, gives individuals a protocol for proposing terms to websites as first parties. MyTerms could become the concrete mechanism through which SEDI’s duty of loyalty requirement, essentially fiduciary obligations to identity holders, are expressed and enforced. I'm at VRM Day before IIW, and the morning's primary topic is MyTerms, the newly published IEEE 7012 standard.

Summary GUIs are dead—at least for most user experiences. This post describes a BYU capstone project where five seniors built a conversational interface for Manifold using MCP and picos. The result shows how natural language can replace a GUI entirely, letting users create, tag, and manage digital things through dialogue instead of learning a standard graphical user interface. Every winter semester, I like to sponsor a capstone project for BYU computer science seniors.

Summary Agents don't just perform actions; they execute plans where the safety of each step depends on what has already happened. That makes sequencing an authorization problem. This post explores how policy, delegation data, and multi-signature approval can govern the order in which agents receive authority, not just the scope of it. This post is part of a series on using dynamic authorization to control and coordinate AI agents. See the series recap to find other posts in this series.