Stop Using Claude Code Like This (Use Sub-Agents Instead)
🚀 Access ALL video resources & get personalized help in my community: https://www.skool.com/agentic-labs/about?ref=3fd61190e13d426dbf4f3b38adc7de69 💬 My AI voice-to-text software (Wispr Flow): https://wisprflow.ai/r?WALTER67 ☕ Buy me a coffee: https://www.buymeacoffee.com/leonvanzyl 💵 Donate using PayPal: https://www.paypal.com/ncp/payment/EKRQ8QSGV6CWW Learn how to use Claude Code subagents and background agents to build complex applications without running out of context window. This tutorial covers the built-in agents like the Explorer, Planning, and Bash agents, and shows you how to create custom specialized agents for coding, UI design, and code review. You'll discover a powerful wave-based workflow that uses parallel subagents to implement features faster while keeping your main conversation clean. ⏰ TIMESTAMPS: 00:00 Claude Code subagents introduction 00:33 Listing built in agents 01:57 Invoking agents directly with @ symbol 02:15 Running agents in background with Ctrl B 03:12 Using Explorer agent for codebase analysis 04:38 Subagents vs main agent context usage 06:06 Benefits of parallel subagents 06:42 Creating custom agents with slash agents 08:06 Selecting agent model Opus Sonnet Haiku 09:54 Context window explained 11:02 Auto compacting and context limits 13:33 Custom UI expert agent results 14:36 Creating coder and code reviewer agents 17:06 Practical subagent workflow overview 17:23 Using planning agents in parallel 20:30 Creating implementation plan with phases 22:48 Parallel agents for documentation 23:39 Wave based implementation strategy 25:28 Orchestrating coder and reviewer agents 27:17 Final code review with multiple agents 28:36 Testing the completed todo app 30:43 Full application with authentication #claudecode #agenticcoding #vibecoding
