My Exact Claude Code Setup Still Allowed After Ban

🛡️ Grab the full source + join Creator Magic Premium → https://mrc.fm/premium (use code TANK for your first 3 months at a discount) I built my own mission control for Claude Code and it's the one setup that's still allowed after the bans and billing chaos this year. I can spin up as many agents as I want, each one sandboxed, and watch every single one work live in the browser. It runs entirely on my own server. No GitHub, I use Forgejo for self hosted git. Nothing leaves the building except the calls to Anthropic. In this video I show you exactly how it works, why it's on the right side of Anthropic's terms (April's cutoff, and the June 15 programmatic credit change), and then I get the dashboard to build itself a brand new feature on camera - flattening a whole repo into one text file. I also run live deep research with sub agents, drag and drop screenshots into Claude Code, and merge real pull requests into my own self hosted git. A quick, honest heads-up: every agent runs with permissions skipped, because I like to live dangerously. It's powerful and a little bit dangerous, so run it isolated, locked down, and never as root. The full source is in my premium community as a snapshot, as is, with no support - bring your own Claude (Max, Pro, whatever you're on). 📦 Get it here → https://mrc.fm/premium (code TANK) 🔗 The terms, if you want to read them yourself: - Anthropic legal & compliance: https://code.claude.com/docs/en/legal-and-compliance - The June 15 programmatic-credit announcement: https://x.com/ClaudeDevs/status/2054610152817619388 If you want me to keep building features on this - or if it's just a fun toy I'm enjoying this week - let me know in the comments and I'll do a follow up. If this saved you from the great Claude subscription panic of 2026 throw a like! 00:00 Introduction: Mission Control for Claude Code 00:34 Starting a New Task & UI Overview 01:16 Anthropic Terms of Service & Terminal Hooks 02:49 Why Build "Tank" & Moving Away from GitHub 03:52 Markdown To Do List Integration 05:10 Creating a Repo Flattening Feature 06:57 Multitasking & General LLM Queries 08:35 Real Time Usage Tracking & Database Storage 09:59 Persistent Sessions with tmux & Drag and Drop Images 11:39 Completing & Testing the Repo Flattening Feature 13:23 Visual Input Testing with Screenshots 15:11 Project Icon Customization 15:36 Security Warnings, Community Access & Wrap up