Luis Arnauth’s Post

At the Xebia XKE this week, we spent a session on SKILL.md files for GitHub Copilot. The room was split. Some had never seen the format. Others were already shipping skills in client repos, with a couple of sharp examples to show for it. The one that got the loudest reaction was grill-me by Matt Pocock. Barely ten lines. It tells the agent to interview you relentlessly about a plan, walk down each branch of the decision tree, and ask one question at a time. A colleague had been pointing it at Microsoft Advanced Specialization artifacts and letting Copilot play assessor for an hour. Good enough that I wanted to write the longer version down. https://lnkd.in/eX_P9sTJ The post covers: - What a SKILL.md actually is and how the index, match, load loop works - Why skills beat one giant copilot-instructions.md - How skills differ from MCP, prompt files, and custom agents - A working Azure Monitor KQL skill you can drop into any repo today - The npx ecosystem forming around skills (including 1000+ ready-made ones) - How to use /create-skill in Copilot to generate one from a conversation And yes, Claude Code follows the same Agent Skills standard.

Running Microsoft Entra Proof‑of‑Concepts (POCs) often involves a lot of repetitive setup, manual validation, and trial‑and‑error before you can really start evaluating what matters. We created Entra‑POCAdvisor to help make that process a bit easier by providing a reusable foundation for teams working on Microsoft Entra pilots and implementation assessments. With this project, you can: ✅ Speed up POC readiness  ✅ Reduce manual validation effort  ✅ Improve consistency across implementations  ✅ Identify potential gaps earlier in the process  ✅ Focus more on evaluating outcomes vs. configuring environments We’re also leveraging GitHub Copilot Skills within the project to help automate and guide key parts of the POC workflow — and explore how AI‑assisted development can support identity scenarios. If you're running Entra pilots today (or planning to), we’d love for you to check it out, try it, and share feedback or contribute. Take a look here:  https://lnkd.in/eEqnPbix Looking forward to learning from how others in the community are approaching Microsoft Entra POCs today 🙌 Andres Canello Wendy Badilla Víctor H. Vargas Martin Coetzer #MicrosoftEntra #GitHubCopilot #OpenSource #Identity #DeveloperCommunity

Found a great resource for those on a GitHub Copilot journey - no matter what UI you put on top - gravity always seems to pull back to the CLI/"Command Line" - the UI that's stood the test of time :) - Quick Start - First Steps - Context and Conversations - Development Workflows - Create Specialized AI Assistants - Automate Repetitive Tasks - Connect to GitHub, Databases & APIs - Putting It All Together https://lnkd.in/ec4hj5Bg #GitHub

From Microsoft Mission Critical Blog articles, Getting Started with GitHub Copilot SDK, by anishekkamal "GitHub Copilot has been a staple in developer workflows for a while — it suggests code, completes functions, and generally keeps you from looking..." https://lnkd.in/eJ_Cn2Ym

S08E17 - Context Is Everything: Getting the Most from GitHub Copilot with Joydip Kanjilal Software architect and Microsoft MVP Joydip Kanjilal joins Jamie to discuss GitHub Copilot and AI-assisted development — covering what Copilot actually is, why context is everything when prompting it, and the governance and training considerations for teams adopting it.

🚀 GitHub Copilot CLI is here — and it’s not just autocomplete anymore. This is agentic AI inside your terminal. It doesn’t just suggest code… it plans, executes, tests, and even ships. Here’s the simplest way to understand it 👇 ⸻ 💡 What it actually does Think of it as a coding assistant that can: • Read your repo • Plan changes • Write code • Run tests • Open PRs 👉 All from your terminal. ⸻ ⚙️ Get started quickly Install → Login → Run copilot That’s it. You’re in. ⸻ 🧠 Make it smarter (important) Add a file: .github/copilot-instructions.md Define your: ✔️ Tech stack ✔️ Naming conventions ✔️ Code standards Now it works your way, every time. ⸻ 🔄 Two powerful modes • Plan Mode → Think first, ask before acting • Autopilot Mode → Executes everything end-to-end ⚡ Pro tip: Start with Plan → Move to Autopilot ⸻ 🤖 Switch models anytime Use /model mid-session No restart. No friction. ⸻ 🔗 Works natively with GitHub You can literally say: 👉 “Review PR #42” And it will: • Analyze code • Detect issues • Suggest fixes ⸻ ⚡ Parallel execution with /fleet Run multiple tasks at once No waiting. Just results. ⸻ 🔌 Extend with MCP servers Connect: • APIs • Databases • Internal tools One config → unlimited capability ⸻ 🎯 Try this prompt: “CI is failing on integration tests. Find the root cause, fix it, run tests, and open a PR if it passes.” 👀 Then just watch what happens. ⸻ 💭 We’re moving from writing code… to managing outcomes. Are you ready to let AI handle your dev workflow?

The Azure Office of the CTO Incubations team recently used GitHub Copilot in a way that I think many engineering teams will find useful: identifying and fixing discrepancies between code and documentation and ensuring that tutorial documentation is complete and accurate. The team behind Drasi, our open-source data change processing platform, faced a common challenge. As features evolve rapidly, documentation can easily fall out of sync. By leveraging GitHub Copilot’s ability to reason across both source code and markdown files, they were able to automate the discovery of these documentation bugs. It’s a great example of using AI not just to write new code, but to maintain the integrity and clarity of an entire project. You can read the full breakdown of their workflow and the results on the Microsoft Open Source blog: https://lnkd.in/gXx33VXh

This is really great, documentation is key to everything and if we can find the gaps and bugs as we all rapidly make changes, improvements and add new features, it ensures we have greatly and up to date documentation.

Great use of AI to solve a everyday issue with keeping everything in sync between code and documentation. How many times have inherited code and the documentation is several versions behind?

GitHub Microsoft Copilot is now restricting signups and resources on all plans, mostly because of long-running parallel computing initiated by stuff like OpenClaw and Geoffrey Huntley's very useful RalphWiggins-style development processes. They're just the latest in a long conga line of similar companies doing the same things. The austerity ahead of us is going to cripple or destroy dev shops and business models that bet the farm on essentially unlimited AI access. We've progressed so far with AI that we've gone back in time to the mainframe timeshare computing model. It was fun while it lasted! https://lnkd.in/e_aBJZv7 #AI #Development #GitHub #Copilot #Microsoft