Extensions

Your Agent Doesn't Know It's in a Container

There's a quiet little failure mode that shows up the moment you bring an AI coding agent into a Dockerized project. You ask it to run the migrations. It cheerfully fires off php artisan migrate on your host machine — where there isn't any PHP, or worse, where there's a stray local MySQL that has nothing to do with your actual app. Nothing happens. Or something happens, which is somehow worse.

Seeing What You Ship

There's a particular dance that terminal-based development teaches you. You ask the agent to nudge a button's padding, or make a gradient less aggressive, or fix that weird thing happening to the heading at tablet width.

Letting Agents Work in Shifts

Or: why one tired agent is worse than five focused ones

Four releases, one day, a lot of small joys

There's something satisfying about a cluster of small releases. No grand reveal, no version bump that demands a blog post all its own — just steady improvements, each one earning its keep. Elyra shipped v0.5.5 through v0.5.8 in a single day, and the changes range from invisible plumbing to features you'll actually type out loud.

Scaffolding with the Real Thing: Laravel Starter Kits in Elyra

There's a familiar little ritual at the start of every Laravel project. You run laravel new, pick a starter kit, brew a coffee — and then spend the next hour squinting at unfamiliar files, trying to work out how the thing actually fits together.

Splitting the Work: How Subagents Let You Think in Roles

Most conversations with an AI coding assistant are just that — one long conversation. You describe what you need, the agent pokes around in your files, writes a bit of code, and you both move on with your day. For small tasks, that's perfectly fine. But once a job starts pulling in several concerns at once — is it correct, is it secure, are there tests, does it fit the architecture?

A Quieter Way to Build Slides: Anna.js and the Elyra Extension