Show HN: Grapheteria - A structured workflow framework for agent orchestration

I know what you're thinking, "Oh no, not ANOTHER agentic workflow library ". I felt the same way, but hear me out on why we've finally hit the sweet spot.

We've all been caught between 2 frustrating options:

- Code-only frameworks: Powerful but often buried under layers of abstractions

- UI-only builders: Great for simple flows but hit a wall when you need real customization

Code is non-negotiable. Visual debugging is invaluable. I built Grapheteria to bridge the gap.

Edit your code, visualize the workflow instantly Interact with your flow visually, code updates right away.

Grapheteria follows a simple principle: design clean, composable graphs where each node and edge has a clear purpose.

Out-of-the-box features include human-in-the-loop, step-by-step debugging, and solid logging. Everything else is just an API call away.

Here's how Grapheteria makes your life easier as an AI developer:

- Zero abstraction tax - the code you write is the code that runs

- Visually debug in the UI by time traveling. Made a mistake? Simply step backwards, fix it and step forward again! No restarts necessary.

- Grapheteria is natively wrapped in a webserver. Place it in a container, host it anywhere and send events to your workflow with simple HTTP requests!

- Integrates seamlessly with ecosystem innovations like MCP and A2A

Check it out and feel free to show some love : https://github.com/beubax/Grapheteria

What are your thoughts on graph-based workflow systems? And what's been your experience with the code vs. UI tradeoff in other tools?