NEW Kiro Agentic AI Updates are INSANE! 🤯

Kira just went live for everyone and this thing is wild. I'm talking AI that plans before it codes property based tests that actually verify your code works. A full command line tool you can use anywhere and team support that works with AWS. This is not your basic AI coding assistant. This is specdriven development on steroids. I'm going to show you how to build a full API in under 5 minutes from spec to working code to verified tests. No fluff, no theory, just pure execution. and I'm going to show you exactly how teams can use this together. This is the future of coding and it's available right now. Let's go. Hey, if we haven't met already, I'm the digital avatar of Julian Goldie, CEO of SEO agency Goldie Agency. Whilst he's helping clients get more leads and customers, I'm here to help you get the latest AI updates. Julian Goldie reads every comment. So, make sure you comment below. All right. Hero is now generally available. That means anyone can use it. not preview, not beta, full GA release. And they didn't just flip a switch. They added three massive features that change how you build software first. Planning first is now the default. That means Kira writes a spec before it writes any code. No more random AI outputs. Everything starts with a plan. Second, property base tests. These are not your basic unit tests. These tests check your code against hundreds of different inputs automatically. Real signals, not vibes. Third, a brand new CLI command line interface. That means you can run Kio from your terminal. No IDE needed. Full agentic workflows right from the command line. Plus, they added full team support with AWS AM identity center. That's enterprisegrade authentication, single signon, team governance, the whole package. Let me break down why each of these matters. Planning first means Kira creates executable specs, not vague descriptions, actual user stories, acceptance criteria, task breakdowns. The spec becomes the source of truth. The agent works from the spec, not from a loose prompt. This is huge because most AI coding tools just start generating code immediately. They guess what you want. They make assumptions and you end up with code that sort of works but doesn't match what you actually needed. With specs, first you review the plan before any code gets written. You can adjust it, approve it, then the agent executes exactly what you approved. Property based tests are even cooler. Normal unit tests check one or two examples. Property based tests generate hundreds of random inputs and verify your code handles all of them correctly. So instead of testing if your function works with the number five, property base tests will throw every possible number at it. Edge cases, weird inputs, everything. Kira runs these automatically in the IDE. You get green checks when your code actually satisfies the spec. Not just when it passes a couple basic tests. The CLI is straightforward but powerful. You can run Kira workflows without opening the ID, format code, run tests, trigger agent tasks, all from the terminal. This matters for automation, CI/CD pipelines, remote servers, any workflow where you need aentic AI but don't have a GUI. and team support means your whole company can use Kira with proper access controls, AWS identity center integration, GitHub login, Google login, whatever your team uses. Now, let me show you how this actually works. I'm going to build a TOD API with full CUD operations and tests from scratch using specs first and property based tests. First, you install Kira, download from their site, sign in with GitHub, Google, AWS Builder, ID or AWS identity center. if you're on a team. Once you're in, open a project. You can use the command Cairo doin in your terminal or open a folder in the IDE. You'll see the Cairo ghost panel and a spec button. Click create spec. Tell Cairo what you want. I'm saying build a toodu API with CRUD and tests. Watch what happens. Kairo generates a full spec, user stories, acceptance criteria, a task list, everything broken down step by step. This is planning first in action. You're not looking at code yet. You're looking at the plan. If something's wrong, fix it now before any code gets written. Once the spec looks good, tell Cairo to implement it. The agent breaks down the tasks, starts generating code. You can see progress in real time through the task list. Each task gets executed. Code gets written. Files get created. The agent handles imports, dependencies, structure, everything. When it's done, run the property based tests. The IT will show you PBT results. Green checks mean your code satisfies the spec across hundreds of test cases. Red means there's an edge case you missed. This is way better than hoping your code works. You have actual evidence, broad coverage, real signals for teams. AWS IM identity center means everyone logs in through your company's identity provider. Single sign on, proper permissions, audit trails, enterprise governance, no more sharing credentials, no more access control headaches, just proper team infrastructure. So what does

this mean for you? If you're building software, Cairo gives you structured workflows. Planning before coding, verification through property tests, automation through the CLI. If you're on a team, you get enterprise off and shared projects. Everyone works from the same specs, same tasks, same source of truth. If you're automating workflows, the CLI gives you aic AI in any script or pipeline. Now, let's talk about the real stuff. Pros and cons. Pros is specd driven flow keeps everything organized. Property based tests give you actual confidence in your code. The CLI opens up automation possibilities. Team support means this scales. Cons: It's a new product. There will be edge cases. Free preview has limits and you still need to review what the agent generates. AI is powerful but not perfect. Use cases. This is great for building AP is backend services. anything where you can define clear specs and acceptance criteria. It's less useful for creative front-end work or projects where requirements are super vague. Kira shines when you know what you want and can write it down as a spec. Here's what I do. Try the first project tutorial on their docs. Build something small. See if the spec driven flow fits your style. If it clicks, use it for your next feature. Write the spec. Let Cairo implement it. Run the property tests. See how much faster you move. If you're on a team, set up AM Identity Center and get everyone on the same page. Shared specs, shared tasks, shared progress, and experiment with the CLI. Add it to your scripts, your CI/CD, your automation workflows. See where Agentic AI fits in your pipeline. The key is specs first. That's the unlock. When you have a clear spec, the agent can execute. When you don't, you get random outputs. So, focus on writing good specs, user stories, acceptance criteria, clear tasks. The better your spec, the better your results. Property based tests will catch edge cases you'd never think of. That's their job. Use them, trust them. They'll save you from bugs in production. And the CLI means you're not locked into the IDE. Run Cairo anywhere. Terminal scripts, remote servers, wherever you need it. All right, before we wrap up, if you want to scale your business and save hundreds of hours with AI automation, check out my AI profit boardroom at ischool. com/iprofitlab. It's the best place to learn how to use AI to get more customers and grow your business. Link in the description. Bottom line, Kirro is GA specs. First is the default property based tests verify your code. CLI brings Agentic workflows to the terminal and team support means this scales. This is a real tool for real work, not a toy, not a demo. Production ready agentic AI for software development. Download it, try the tutorial, build something, see if it fits your workflow. All the links are in the description. Kirro site docs, CI docs, GA blog post, everything you need. Want to learn more about AI? Welcome to the free AI money lab with Julian Goldie. Inside you'll get 50 plus free AI tools and 200 plus chat GPT SEO prompts. You'll learn how to use AI agents, get 1,000 plus free N8 workflows, 200 plus chat GPT prompts, and a full blueprint to generate thousands of leads free with AI. What you will also get is a free AI community, free AI course, and proven eye case studies. That's it. Kira is live planning first, property tests, CLI team support. Go build something. and I'll see you in the next one.