I’ve spent a lot of time experimenting with different ways to extend AI agents—system prompts, MCP servers, tool definitions—and they all work, but they don’t scale cleanly. Agent Skills finally feel like the right abstraction. Agent Skills are an open standard...
I’ve been building conversational agents with Azure Logic App Agent Loops since the feature first landed. Early on, the biggest limitation was how tightly tools were coupled to individual agents. Every tool lived inside the agent loop, which meant duplication, harder...
When selecting an AI image model for your application, you are usually forced to choose between quality and control. Proprietary models offer great images but lock you into their cloud. Open-source models offer control but can be a headache to host and quality is not...
When I need flexible message routing in Azure without duplicating infrastructure, Azure Service Bus Topics are usually my first choice. They solve a very specific problem: letting multiple consumers react differently to the same message without forcing the publisher...
I found that Logic Apps Consumption with AgentLoop is one of the fastest ways to get a practical AI agent running—especially if you’re experimenting or running low-volume workloads. In this post, I’ll walk through why I chose Consumption, how the agent is structured,...
Recent Comments