<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Github-Copilot on Agentic Engineering in Action (agentprose demo)</title><link>https://vibeprogrammer.github.io/agentprose-demo/tags/github-copilot/</link><description>Recent content in Github-Copilot on Agentic Engineering in Action (agentprose demo)</description><generator>Hugo</generator><language>en-US</language><lastBuildDate>Thu, 23 Jul 2026 06:32:27 +0000</lastBuildDate><atom:link href="https://vibeprogrammer.github.io/agentprose-demo/tags/github-copilot/index.xml" rel="self" type="application/rss+xml"/><item><title>Giving AI agents a real computer</title><link>https://vibeprogrammer.github.io/agentprose-demo/posts/giving-ai-agents-real-computer/</link><pubDate>Thu, 23 Jul 2026 06:32:10 +0000</pubDate><guid>https://vibeprogrammer.github.io/agentprose-demo/posts/giving-ai-agents-real-computer/</guid><description>&lt;h3 id="how-i-built-sandcastle-on-the-new-github-copilot-provider-for-microsoft-agent-framework--and-what-the-feature-actually-unlocks"&gt;How I built Sandcastle on the new GitHub Copilot provider for Microsoft Agent Framework — and what the feature actually unlocks&lt;/h3&gt;&#10;&lt;blockquote&gt;&#10;&lt;p&gt;▶️ Live demo: &lt;a href="https://sandcastle.isainative.dev/"&gt;https://sandcastle.isainative.dev&lt;/a&gt;&lt;/p&gt;&#10;&lt;/blockquote&gt;&#10;&lt;p&gt;Most &amp;ldquo;AI app&amp;rdquo; demos boil down to the same thing: a prompt goes in, a wall of text comes out, and you still have to copy it, run it, fix the three things that don&amp;rsquo;t compile, and host it. The &lt;a href="https://learn.microsoft.com/en-us/agent-framework/agents/providers/github-copilot?pivots=programming-language-python"&gt;GitHub Copilot provider for Microsoft Agent Framework&lt;/a&gt; flips that: instead of returning text about code, an agent gets a real, sandboxed computer — it can run shell commands, read and write files, fetch URLs, and call MCP servers — and it uses that computer to actually build and run the thing you asked for.&lt;/p&gt;</description></item><item><title>You can't tune what you can't see: prompt‑cache observability for GitHub Copilot with OpenTelemetry</title><link>https://vibeprogrammer.github.io/agentprose-demo/posts/you-cant-tune-what-see-promptcache-observability-github/</link><pubDate>Wed, 08 Jul 2026 12:42:31 +0000</pubDate><guid>https://vibeprogrammer.github.io/agentprose-demo/posts/you-cant-tune-what-see-promptcache-observability-github/</guid><description>&lt;p&gt;Open VS Code, ask Copilot Chat one question, and if you crack open the debug view you can watch prompt‑cache tokens tick past for that single call. That&amp;rsquo;s exactly enough visibility for one developer, on one machine, for one prompt. It stops being enough the moment you want to answer a sharper question:&lt;/p&gt;</description></item><item><title>Design, but with tests: an engineering approach to UX with Impeccable</title><link>https://vibeprogrammer.github.io/agentprose-demo/posts/design-tests-engineering-approach-ux-impeccable/</link><pubDate>Wed, 01 Jul 2026 14:50:55 +0000</pubDate><guid>https://vibeprogrammer.github.io/agentprose-demo/posts/design-tests-engineering-approach-ux-impeccable/</guid><description>&lt;p&gt;Impeccable — now built into the GitHub Copilot app — treats design quality the way we treat code quality. This is what that looked like on a real app that was already shipping.&lt;/p&gt;&#10;&lt;p&gt;We have version control, code review, linters, profilers, and prioritized backlogs for our code. For the design of the things we build, most of us have &amp;ldquo;looks good to me.&amp;rdquo; This is the story of what happened when I closed that gap — by pointing a skill at my aurora app that brings an engineering discipline to UX.&lt;/p&gt;</description></item><item><title>The model is a commodity. The harness is your advantage. A hands-on experiment with GitHub Copilot app's Canvas</title><link>https://vibeprogrammer.github.io/agentprose-demo/posts/model-commodity-harness-your-advantage-hands-on-github/</link><pubDate>Tue, 30 Jun 2026 12:29:09 +0000</pubDate><guid>https://vibeprogrammer.github.io/agentprose-demo/posts/model-commodity-harness-your-advantage-hands-on-github/</guid><description>&lt;p&gt;The teams getting the most out of AI agents are not the ones with the best model. They are the ones building a custom layer on top of it. This is a short, hands-on look at one new surface for that layer - GitHub Copilot app&amp;rsquo;s Canvas - learned by building a small internal tool end to end.&lt;/p&gt;</description></item></channel></rss>