I still make games. Aviquill is the tool I open before the engine, not instead of it. Most of the actual thinking for a project happens here, on a canvas, before a single line of code or a single asset gets touched.
One board per project
Each game gets its own canvas. Top-level blocks for the pillars — core loop, art direction, target scope — with groups branching off for systems, levels, and UI. When a project grows past "a few ideas," I split it into clusters instead of scrolling further out, the same instinct I'd use to keep a Blueprint graph from turning into spaghetti.
Having everything on one board means I can zoom out and see the whole shape of a project at once — which systems are fleshed out, which are still a single empty block with a title, which ones haven't been touched in weeks. That view is hard to get from a folder full of docs.
Mechanics as blocks, not docs
Instead of a design doc nobody rereads, every mechanic gets its own block: what it does, why it exists, what it touches. I link related mechanics together the same way I'd wire nodes in a Blueprint graph — so dependencies are visible instead of buried in a paragraph three pages into a doc nobody opens twice.
If a movement mechanic depends on a stamina system, that connection is a visible line on the canvas, not something I have to remember or cross-reference. When I change one, I can see immediately what else sits downstream of it before I touch any code.
Bugs and playtest notes live next to the system they break
A bug found during playtesting goes straight onto the block for the system that caused it, not into a separate tracker I'll forget to check. Same with feedback — if a playtester says a level feels too long, that note goes directly onto that level's block, not into a spreadsheet that gets stale the moment the session ends.
By the time I sit down to fix something, the context is already sitting right next to it: the original design intent, every related mechanic, and whatever broke last time I touched this part of the system.
Why this beats a wiki or a tracker
A wiki wants everything in a tree. A tracker wants everything to be a ticket with a status. Game design doesn't actually work that way — ideas are messy, half-formed, and connected to other ideas in ways that don't fit neatly into either shape.
A canvas doesn't ask me to decide the hierarchy up front. I can place things where they make sense, connect them when a relationship becomes obvious, and reorganize without losing anything when the project's structure inevitably changes halfway through development.
It's not a replacement for the engine. It's the layer above it — where the reasoning lives before any of it becomes code.