We are breaking down the massive AI news surrounding the recent security breach. This episode discusses the unprecedented anthropic leak, which exposed the claude source code and led to a significant claude code leak, impacting half a million lines of code. We explore the implications of this event and what it means for cyber security in the AI world.

💡 Sponsor: Huge shoutout to Auth0!

Secure AI agents, humans, and whatever comes next

https://auth0.com/

✅ If you made it to the end, comment: “Claude Code Undercover” so we know who is a real one!

https://www.commityourcode.com/

Come to CYC!

Get The Developers Guide To AI here!

http://developersguide.ai/

SITE https://www.programmingpodcast.com/

Stay in Touch:

📧 Have questions for the show? Or are you a business that wants to talk business?

Email us at [email protected]!

Danny Thompson

https://x.com/DThompsonDev

www.DThompsonDev.com

Leon Noel

https://x.com/leonnoel

https://100devs.org/

📧 Have questions for the show? Or are you a business that wants to talk business?

Email us at [email protected]!

Subscribe if you want honest, no-hype conversations about where software development is heading in 2026.

Chapters

00:00 - Intro: The Largest AI Leak in History

01:47 - By the Numbers: Secret Variables, Commands & Build Flags

03:17 - Coordinator Mode & The Sub-Agent Hierarchy

06:56 - UDS Inbox & Auto Dream: The Future of AI Memory

10:24 - Auth0

10:51 - Uncovering Hidden Slash Commands (/dream, /bug_hunter)

13:41 - Exploring Claude's New Voice Mode & Dictation Tools

16:26 - "Undercover Mode": Ghostwriting Open Source Commits

18:27 - The Not-So-Fun Leaks: 6x Fast Mode Markups & HIPAA Limits

23:34 - Breaking Down Claude's Step-by-Step Query Loop Architecture

26:59 - OpenAI's Massive $150 Billion Valuation Announcement

28:03 - Ask Danny: The Ultimate Post-Coffee Chat Follow-Up Strategy

35:27 - Outro & Wrap-Up