Plan, Specify, and Implement with Spec Kit

Spec Kit is an open-source toolkit that allows you to focus on product scenarios and predictable outcomes instead of vibe coding every piece from scratch. Den shows off what it is, how to get setup, and how to get started with spec drive development with GitHub Copilot. Join the conversation on the Microsoft Foundry Discord: https://aka.ms/AIDevDays/Discord Sign up for an Azure Free Trial: https://aka.ms/AzureFreeTrialYT Get hands-on with today's VS Code lab repo: https://aka.ms/AIDevDays/LabDay2 Catch up on-demand: https://aka.ms/AIDevDays/OnDemand Continue learning next week with AI Apps & Agents' global sessions: https://aka.ms/AIDevDays/Next Register for the January AI Dev Days Hackathon: https://aka.ms/AIDevDays/Hackathon Chapters: 00:00 - Intro: James and Dan Kick Off with Spec-Driven Development 00:27 - Why Planning Matters: From Vibe Coding to Structured Specs 01:41 - What Is Spec-Driven Development and Why It Scales 03:52 - Context Is King: Why Specs Improve AI-Generated Code 05:33 - Specs vs. Custom Instructions: How They Work Together 06:41 - Inside SpecKit: Prompts, Scripts, and GitHub Integration 08:34 - Demo: Setting Up SpecKit with Specify CLI 12:03 - Adding SpecKit to Existing Projects 14:03 - Creating a Constitution: Non-Negotiable Project Principles 18:01 - Building Functional Specs and Clarifying Requirements 21:02 - Clarify Mode: AI Questions You Didn’t Think to Ask 25:33 - Technical Plan Stage: Adding Tech Details and Dependencies 27:01 - Choosing the Right AI Model for Spec Workflows 28:53 - Branching, Variations, and Git Workflows for Specs 31:02 - Breaking Down Tasks and Automating Implementation 32:01 - Reviewing Generated Artifacts: Plan, Data Models, and Research 34:54 - Scaling SpecKit for New Features and Existing Apps 35:17 - Wrap-Up: Why SpecKit Changes How We Build Software