Loading video player...
Spec-driven development is everywhere, but if you're using Claude Code, you already have it built in. Plan Mode has evolved into a full spec-driven workflow with local storage, subagents, and task dependencies. In this video, I break down how Plan Mode works: the four phases (Exploration, Design, Review, Final Plan), the January 2026 updates, including local plan storage and task dependency tracking, and the built-in subagent orchestration that external tools can't match. I also demo adding a real feature to show the complete workflow. Whether you're using OpenSpec, Spec Kit, or considering them, this video shows you what Claude Code can do natively. For small to medium features, you might not need anything else. ā±ļø TIMESTAMPS 0:00 Introduction 0:41 The Problem with AI Coding 1:03 Plan Mode: The Native Alternative 2:12 The Four Phases 3:04 The January 2026 Updates 4:45 Built-in Subagents 5:06 Demo: Adding a Feature 6:19 Why Not External Spec Tools? 6:47 When to Use Plan Mode 8:16 Conclusion š LINKS & RESOURCES Claude Code ā Plan Mode https://code.claude.com/docs/en/common-workflows#use-plan-mode-for-safe-code-analysis OpenSpec https://github.com/Fission-AI/OpenSpec Spec Kit (GitHub) https://github.com/github/spec-kit Join the Discord Community https://discord.gg/PKgAUuGnGG Follow on X/Twitter https://x.com/thegraytcat š± THE GRAY CAT AI meets code. Building next-generation software with modern AI tooling. Covering AI IDEs, Claude Code, MCP integrations, and TypeScript/Node.js development. New videos every week. Subscribe and hit the bell!