All videos
0:00 / 0:00
ai

3 Secret Ways Claude Code’s Creator Actually Uses It

Nate Herk | AI Automation12 August 2026Watch on YouTube

Part of series

Ep. 6 · Claude AI SEO-systeem

Complete workflows voor het automatiseren van SEO-contentcreatie en websitegroei met Claude AI en de Hermes Agent.

View the series

What you'll learn

  • Simplify old AI instructions and context, because Opus 5 can actually perform worse when constrained by legacy prompts
  • Prompt at a higher level without micromanagement: set goals and verification methods instead of step-by-step instructions
  • Poorly configured hooks and custom prompts from older models can now hinder Claude Code rather than help it

Frequently asked questions

Why does Claude Code feel worse since Opus 5?
Older instructions, hooks, and CLAUDE.md guidelines built for previous models can now work against Opus 5. Context designed to help can now limit the capabilities of the stronger model.
What are the three changes you should make for better Claude Code usage?
Simplify old instructions, prompt at a higher level, and give the agent a way to verify its own work. The goal is not to delete everything, but to keep necessary context and stop micromanaging.
Should you always delete all old prompts and skills?
No, it is not about deleting everything. The advice is to selectively adjust your context and not blindly copy advice from others that may not fit your specific situation.

Topics

claudeai-automationpromptingcode-generationllm-optimization

Description from the channel

My playbook for growing a $1M AI agency: https://app.aiautomationsociety.ai/opaa-ads-optin My FREE resources: https://www.skool.com/ai-automation-society/about?el=claude-delete&hcategory=youtube-videos&utm_campaign=free-group My Tools💻 FREE MONTH voice to text: https://get.glaido.com/nate Code NATEHERK for 10% off VPS (annual plan): https://www.hostinger.com/vps/claude-code-hosting Check out the YC interview with Boris: https://youtu.be/qyPCVqFUyDo?si=4OVk-AqSS62wr7PC Claude Code may feel worse with Opus 5 because the skills, hooks, and `CLAUDE.md` instructions built for older models can now get in its way. I break down Boris Cherny's advice, show what happened when I tested Claude with and without my AI OS context, and share the three changes I'm making: simplify old instructions, prompt at a higher level, and give the agent a way to verify its own work. The goal is not to delete everything. It is to keep the context Claude needs and stop micromanaging how it gets there. Sponsorship Inquiries: 📧 nate@smoothmedia.co Connect with me: https://www.linkedin.com/in/nateherkelman/ https://x.com/nateherk https://www.instagram.com/nateherk/ TIMESTAMPS 0:00 Why Claude Feels Worse 0:23 Anthropic Deleted 80% 2:05 Delete Your Old Instructions 3:49 My No-Skills Test 5:56 Don't Blindly Copy Advice 7:18 Unhobbling the Model 8:46 Goals and Verification 11:31 Final Thoughts