graph TD
A[🤖 AI Agents] --> B[⚡ n8n Brain]
B --> C[🏗️ DevOps Arsenal]
C --> D[🏠 Chaos Homelab]
A --> E[Claude API<br/>💬 Chat Master]
A --> F[OpenAI<br/>🧠 Think Tank]
B --> G[Webhooks<br/>📡 Signal Tower]
B --> H[Automation<br/>⚙️ Magic Box]
C --> I[Kubernetes<br/>☸️ Pod Herder]
C --> J[Terraform<br/>🏗️ World Builder]
Where AI meets infrastructure... and occasionally explodes 💥
Automate smarter with AI + n8n. 🤖⚡
Short, copy-paste tutorials and real DevOps tips from a homelab-obsessed engineer who thinks sleep is overrated. 🏠🧪☕
Mission: We break things so you don't have to — and automate so we can break better next time. 🔁🔥
- 🧩 n8n workflows & AI agents (Claude, OpenAI, MCP)
- 🔗 Webhooks, APIs, scraping, observability that actually work
- 🛠️ Practical DevOps: GitOps, Kubernetes, Terraform, Ansible, Proxmox
- 🎁 Free templates + project code (because sharing is caring)
> cd /ai-ops-lab
> ls
📁 beginner-friendly/ 📁 expert-level/ 📁 probably-cursed/
📁 youtube-demos/ 📁 emergency-fixes/ 📁 coffee-required/
> cat warning.txt
⚠️ Some automations may achieve consciousness. We're not responsible for any AI uprisings.
> ./run_workflow.sh --help
Usage: ./run_workflow.sh [--pray] [--backup-first] [--blame-dns]
--pray Increases success rate by 12%
--backup-first For the cautious (recommended)
--blame-dns When all else fails (works 60% of the time, every time)
Did these workflows prevent a rage-induced server restart?
Because caffeine is the only dependency that never breaks. ☕️⚡
P.S. No homelabs were harmed in the making of these workflows. (Okay, fine, three servers, two network switches, and one very confused UPS. But hey, that's how we discovered that "turning it off and on again" works even better when automated!) 💾🔥⚡
P.P.S. If your automation gains sentience and starts ordering pizza at 3 AM, that's a feature, not a bug. We call it "predictive hunger management." 🍕🤖