A platformer maker with no ceiling on it

Search this genre and you find dozens of tools with a character, a jump, some enemies and a menu of behaviours. They are honest fun for an afternoon. The trouble arrives the moment your game wants something the menu does not have: a grappling hook, a gravity flip, an enemy that mimics your last three inputs.

At that point you are stuck, and stuck permanently, because the tool was built on a fixed list of possibilities. There is no advanced mode. The ceiling is not a skill curve, it is a wall.

Flockbay has no such list. Behaviour is code, and code is written on demand, so the question is never whether the feature is supported.

You describe the mechanic. Director 1 writes the GDScript inside your Godot project. If it is wrong you say what is wrong and it changes. The novelty of your idea has no bearing on whether it can be built, because nothing is being looked up in a catalogue.

That extends past mechanics. Menus, saves, checkpoints, a level select, a speedrun timer, an accessibility option that halves the game speed — all of it is just code someone has to write, and here that someone is not you.

The one thing you must supply is judgement. You play it, and you decide whether it is right. You will do that many times, and it is the real work.

The builder is where you spend most of your time. Placing platforms, checking sight lines, standing where the player will stand and asking whether the next jump is legible before you make it.

Which builder that is gets decided before the project exists. Making one asks 2D or 3D, shown as two pictures, with no default and nothing created until you pick. Pick 2D and what opens is already a game: sky, a band of scenery along the horizon, ground with a ledge cut into it, a platform floating off to one side, and a person who walks, falls and jumps under the keys. You are seconds in and already arguing with the jump, which is where this genre's work happens.

Platforms are painted rather than picked out of a palette. You drag to lay ground and drag to erase it, and the tiles, the edges and the collision all fall out of which cells are ground, so the ledge you just dragged is one the player can land on and climb to before you have asked anyone for anything. Paint the same cells in a different order and you get the same level, and a rebuild never reshuffles the edges you already accepted.

Pick 3D and you get that side's own tools instead: terrain, water and weather, so a platformer through a mountain pass with fog rolling in is an ordinary thing to build here rather than an ambitious one. The flat world has a shorter list — ground, what is beyond it, the person, and the things you place — because water and weather have nothing to describe without depth. Shorter is the honest scope of a flat world, not a smaller advantage.

It is a desktop app for Mac and Windows. There is no browser version, no phone editor, no cloud workspace. Godot ships inside it and your game runs locally on your own hardware.

Building needs an internet connection because Director 1 and asset generation come over the network. Playing what you built does not, and there is no offline building mode.

Free to download, and the builder itself is free — levels, cameras, your own art, playtesting and publishing. Indie at $10 a month unlocks the AI assistance, and the free plan includes a trial of it. There is no team, per-seat or enterprise pricing, because it is a desktop app.

The most reliable way to begin is small and specific. One movement idea, one room that uses it, and a version you can play within the day. Not a level, not a character, not a menu. One room.

What that gets you is the only information that matters early: whether the idea is fun to do repeatedly. Most ideas are not, and finding out on day one is a gift rather than a setback. The ones that survive are worth the months.

From there it grows in the order that keeps you honest. Second room that varies the idea. Third that combines it with something. If you cannot make three interesting rooms out of a mechanic, it is a mechanic for a section rather than a game, and knowing that early is worth more than any amount of planning.

Late. Placeholder shapes are not a compromise in this genre, they are the correct tool for as long as the movement is still changing. A grey box that jumps well tells you more than a beautiful character that does not.

When art does arrive it comes from generation, from the web, or from your own hand, and Flockbay says which route it used rather than blurring it. Geometry built directly in the builder is often the right answer when the style is clean or when it is the honest fastest way to get you playing.

The exception is readability, which is a design problem wearing an art costume. If the player cannot tell which ledges are standable, that is not something to fix after the art pass. Fix it while everything is still boxes.