GitHub Copilot CLI: AI-Powered Terminal

Okay, real talk: I thought Claude Code was just a fancier Copilot. Then I actually used it. This thing doesn't sit around waiting for instructions like an intern on their first day. It moves. Need it to dig through your codebase, run terminal commands, and edit files across your whole project at once? Done. You describe the goal; it maps the route. You're the GPS destination, not the driver. MCP servers are where your jaw drops a little. Plug in external tools, browsers, databases, and APIs, and Claude Code picks them up and uses them like it's always had them. It's not "AI plus tools bolted on." It's AI that actually has a toolbox. GitHub connectors mean it's not hiding in a tab somewhere while your real work happens elsewhere. It's in the PR. It's in the review. It's part of how the team ships, not a side quest. And then there are hooks, which honestly should be talked about way more. Imagine being able to whisper to Claude Code before it does anything: "Check this." "Always do that after," or "never touch this file." Enforce standards. Trigger tests. Build guardrails. It's your workflow, your rules. Claude Code just follows them. Four things. Tools, MCP servers, connectors, hooks. And suddenly you're not just using AI to code faster; you're using it to work smarter. There's a difference. A big one. 🙌 #ClaudeCode #Anthropic #AI #SoftwareDev #DevTools #Automation

GPT-5.5 Is Live in Copilot. This One’s Actually Different. GPT-5.5 just dropped inside GitHub Copilot. Here’s why developers should care. 🧵 Forget chatbots. This model is built for real coding work — multi-step agentic tasks, debugging complex codebases, and running long workflows across tools without falling apart. Early testers describe it as the first model that actually understands how a codebase fits together. Not just autocomplete. Actual reasoning about your code. What’s new: → Best-in-class on complex agentic coding benchmarks → Fewer tokens, same speed, better results → Available across VS Code, JetBrains, Xcode, GitHub Mobile, and more Who gets it: Copilot Pro+, Business, and Enterprise users. The catch? A 7.5× premium request multiplier at launch. It’s not cheap to run — but if it saves you hours of debugging, the math still works. The rollout is gradual. If it’s not in your model picker yet, it will be soon. The bar for AI coding tools just moved. Again. #OpenAI #GitHubCopilot #GPT55 #AIEngineering #DevTools

🚀 Excited to share something I built! I’ve just published my first VS Code extension — PromptCraft AI 🎉 https://lnkd.in/efTDETzn As developers, we often struggle not with coding… but with how to ask the right question to AI. That’s exactly what I wanted to solve. 💡 PromptCraft AI works directly inside GitHub Copilot Chat: 👉 Just type @promptcraft and your rough idea gets converted into a structured, professional prompt. ⚡ Features: Converts vague requests into clear engineering prompts Supports commands like /debug, /refactor, /review Uses your existing Copilot model (no API key needed) Helps you think better before asking AI Example: Instead of: “api failing” You get: ✔️ Task ✔️ Context ✔️ Checks ✔️ Constraints ✔️ Expected Output 👉 This leads to MUCH better AI responses. This was a great learning journey — from idea → design → building → publishing 🚀 Would love your feedback 🙌 Try it out and let me know what you think! #VSCode #GitHubCopilot #AI #DeveloperTools #Productivity #BuildInPublic

At Fountane, we build products fast. That pressure exposed a real problem with AI coding agents. They'd confidently write code for a codebase they barely understood. No warnings, no caveats — just wrong decisions that looked right until they broke something. So I built a fix: a skill you drop into Cursor, Claude Code, or any AI tool that reads markdown. Before your agent writes a single line, it scores itself: — How well does it understand your codebase? — What can it build autonomously right now? — What gaps exist, and what closes them? The real unlock wasn't better prompts. It was knowing the agent's confidence level before giving it work. A 60% understanding score means you're going to spend more time reviewing than building. A 90% score means you can actually delegate. We now run this before any major feature work. It's changed how we structure context, how we onboard agents to new repos, and how we catch blind spots early. Open source. Tool-agnostic. One command to install. If you enjoy thoughtful conversations with people building real products, this could be for you. Apply for an invite → https://lnkd.in/gZdbqS4J Link : https://lnkd.in/dB5Cb9Wp #ProductEngineering #AgenticAI #BuildingInPublic

🚀 Just dropped a new GitHub Selection video featuring 5 open-source projects worth watching this week. These projects are pushing forward AI agents, automation, browser control, and developer workflows in really interesting ways. Featured projects in this video: 🔹 Superpowers – a skills framework for coding agents with real dev workflow support 🔹 MiroFish – a swarm intelligence engine for simulating scenarios and predicting outcomes 🔹 Lightpanda Browser – a lightweight headless browser built for AI agents and automation 🔹 Claude HUD – a live HUD for Claude Code showing tools, context, agents, and progress 🔹 Page Agent – an in-page AI agent that controls web interfaces with natural language Open source keeps moving fast — and these are some of the projects that stood out to me this week. 👀 Which one would you try first? 👇 #OpenSource #GitHub #AI #AIAgents #Automation #DeveloperTools #TechInnovation

Your AI model has a fixed brain. And you're paying for it. It's not dumb. It's actually brilliant. But it's frozen in time — it knows nothing about your current project, your latest sprint, or the code your team pushed this morning. So every single day, your developers are doing this: → Open new chat → Re-explain the entire codebase → Burn tokens on context that should already be there → Repeat tomorrow So I designed something different. What if your model automatically updated itself every time code was pushed to GitHub? Here's the architecture I came up with: 1. GitHub push → webhook fires → diffs get indexed into a local vector store 2. Developers query a small self-hosted model (like Mistral / Phi-3) that already knows the codebase 3. Third-party APIs like Claude/OpenAI/gemini only get called when the local model genuinely can't answer — keeping costs near zero 4. Your code never leaves your own server I call it a Live model — vs the Fixed Brain Frozen model most teams are running today. Has anyone built something like this? What broke? What would you do differently? Drop it in the comments — I'm genuinely curious. #AI #SoftwareEngineering #LLM #DevTools #AIArchitecture #Claude

I’ve been fine-tuning my development workflow lately, and it’s finally reaching a point where the AI doesn't just "help" it actually manages the heavy lifting. If you're curious about how to layer these tools for maximum output, here is the stack I’m currently running: 1. The Research Layer: Gemini + NotebookLM Instead of spending 2 hours digging through dense technical specs, I "interrogate" them. I get the context I need in 10 minutes. It’s like having a librarian who has already memorized every page. 2. The Architect: Cursor Once I have the plan, I use Cursor to gather codebase context. It’s much faster than a standard IDE for mapping out how new features will actually fit into existing code. 3. The Muscle: Claude Code This is where the automation happens. I delegate the coordination and repetitive tasks here. It’s essentially my "Agent" that handles the grunt work while I focus on the big picture. 4. The Gatekeeper: GitHub Copilot Reviewer Before a human ever touches the code, Copilot does a review pass. It catches the "obvious stuff" so my team doesn't have to waste time on trivial fixes during PRs. The result? I’m thinking more and typing less. I’m currently looking at migrating some of this to Obsidian to keep my knowledge base even more organized. How are you all using AI in your dev workflow? Are we at the "autopilot" stage yet, or are we still co-piloting? #SoftwareEngineering #AI #Productivity #GithubCopilot #Claude #Gemini #CodingLife

⚠️ Addictive tech warning for developers. Once you add a 🦆rubber duck to your AI agent pipeline, you’ll start feeling uncomfortable without it. This is exactly what happened to me. I no longer want to rely on a single model’s opinion for important technical decisions, and I definitely don’t want extra manual steps just to get a second perspective. That’s where “Rubber Duck”, an experimental feature in the GitHub Copilot CLI, really worked for me: - enable it with: "copilot --experimental" (Rubber Duck is the 1000th reason for you to switch to terminal-first development) - watch one LLM actively criticise another’s decisions right at the moments where it matters most, pushing towards a better solution - everything happens automatically, no extra friction, no context switching It is a targeted reviewer that steps in at high-value moments such as after drafting a plan, after a complex implementation, and after writing tests before execution. That feels like a very practical way to reduce compounding errors early, especially in long-running or multi-file tasks. So having AI challenge AI has quietly become part of how I build now. Would you trust critical technical decisions to a single model, or is multi-model critique the new baseline for serious AI-assisted development? Ready to try Rubber Duck? I warned you :) More details: https://msft.it/6044Q4Zs2 Morten Stange Bye, Haakon Hasli, Christian Tryti, Else Tefre, Francesco Manni, Jaime De Mora, Martin Woodward, Lee Stott, Christoffer Noring, Daniel Meppiel, Joel Norman, Ömür Sert, Adil I., Sebastien Le Calvez, 🥑 Aaron Powell, Nick McKenna, Burke Holland, Cornelia Bjørke-Hill #GitHubCopilot #GitHubCopilotCLI #CopilotCLI #DeveloperTools #AIAgents #CopilotRubberDuck #msftadvocate

AI isn’t a replacement for thinking, it’s a productivity layer. I use tools like Claude and GitHub Copilot to handle the heavy lifting of boilerplate, SQL drafts, and documentation. This allows me to move faster and save my mental energy for what matters: architecture and logic. The goal isn't just to write code faster, but to spend more time solving the harder problems. #SoftwareEngineering #AI #Productivity #GitHubCopilot #ClaudeAI #CodingLife

git commit -m "I didn't write this" That's the name of our session at Minnebar 20 on May 2nd. Jackson Tomlinson and I are going to build software live (yes, using AI-assisted development) and as the bots are churning tokens we can talk about what we've found that works and what hasn't been as helpful. The point isn't "look how fast or great AI is." The point is that building great software still requires people who understand what and how to build. Architecture, data modeling, user experience, knowing when to listen to a tool, when to lean into it, and when to just blow it up and start over is more important than ever. Expertise matters more now, you just spend less time typing. If you're an experienced developer curious about AI workflows, or a skeptic who wants to see where the edges cut deepest, please join us.  Minnebar 20, May 2nd  https://lnkd.in/gxNaftHR And, of course, you know implicit in all of the above is follow the session link and let the world know you will attend our session.  #minnebar20 #softwareengineering #AI #buildinginpublic