A puzzle game where your tools are completely random

At first glance, Mosa Lina looks like a typical puzzle platformer. The player is dropped into a single-screen room, asked to collect or destroy fruit, and then expected to reach the exit.

The tools seem useful enough: a spell that moves platforms, a phaser that teleports the player through surfaces, and lumpy tentacles that can be spawned at will. One attempt might involve dragging an apple off screen, phasing through a vine, and then realizing there is no way back up.

That failure is the point. The levels are fixed, but the mechanics are completely random. Another attempt might give the player a butterfly teleport, a gravity-reversing rectangle, and the power to pin platforms permanently. A different attempt might hand over an entirely new combination.

Mosa Lina is aggressively random, which means there are no intended solutions in the usual puzzle-game sense. There are no obvious lock-and-key connections, and there is not even a guarantee that the current loadout can finish the level.

The upside is that the player gets to be creative and crafty. They might shoot an apple out of the sky with a spear, then use it as a makeshift platform before it drops away. They might freeze spinning spikes to make a safe bridge. They might jump off a cliff at exactly the right moment, trusting a delayed teleport to return them to safety.

The game turns each room into a strange little improvisation problem. The question is not "what did the designer intend?" It is "what can I make these tools do right now?"

That kind of problem solving recalls immersive sims: games descended from titles such as Ultima Underworld and System Shock, built around powers, tools, magic, interlocking systems, and enemy behavior that let players invent unusual solutions.

A classic example is using wall-mounted mines in Deus Ex as improvised stepping stones to climb a building. The tool was not only a mine. In the right situation, it became a platform.

Mosa Lina's developer, Stuffed Wombat, has described the game as an immersive sim, and also as a response to more modern games such as Deathloop and Dishonored. In those games, abilities can sometimes feel less like open-ended tools and more like keys for specific locks: get super strength, then move the heavy box blocking the alley.

Mosa Lina pushes back against that rigid puzzle-piece mentality. The game gives the player tools, then leaves it to the player to find the solution.

The first clever part is that tools are just tools. Mosa Lina is full of funny, strange, inventive little powers that manipulate the world in interesting ways, but they are not tied to one object or one setup.

A tool does not announce its intended socket. It simply does something, and the player has to find a use for that behavior. Sometimes the best answer is a single power. Sometimes it comes from combining several powers in a way the level never explicitly asked for.

That distinction matters. When an ability is designed only to open a matching door, the player is solving a mapping problem. When an ability has general behavior, the player is solving a systems problem.

The second clever part is physics. Tentacles wobble, pendulums swing, and tools can make platforms move, float, drop, or collide in messy ways.

Games such as Kerbal Space Program and Tears of the Kingdom show how quickly possibilities explode once robust physics enter the design. The designer no longer needs to author every solution directly because the system can produce outcomes from the interaction of forces, objects, and timing.

Mosa Lina compresses that feeling into tiny rooms. The result is slapstick failure, clever recovery, and last-second saves. A typical immersive sim might produce one or two memorable emergent stories in a playthrough. Mosa Lina can create them every few minutes.

The third clever part is randomness. If the developer does not know which combination of tools the player will have, then the developer cannot design one perfect solution. That uncertainty forces the game to become more open-ended.

If a level simply does not work with the current tools, the player can reroll and try again with a new set of mechanics. That might sound like a flaw in a normal puzzle game, but here it supports the whole structure.

It also gives the game a roguelike rhythm. Each run offers a small selection of items from a larger pool, so the player keeps returning to see new combinations and synergies. The appeal is similar to wondering what cards, items, or upgrades will define the next run of a favorite roguelike.

This is not a game for everyone. The player needs to be comfortable with confusion, chaos, wonky physics, and strange presentation. There is little story or structure, so the game can feel more like an experimental toy than a complete traditional adventure.

Some players also do not want to invent their own solutions. They want a crafted logical conundrum with a clear answer, not a systems playground where the answer depends on a random toolset and a lucky bit of physics.

That makes Cocoon an instructive contrast. Cocoon is a handcrafted puzzle adventure about worlds within worlds, where each colored orb is both a tool and a pocket universe. It is almost the opposite of Mosa Lina: no randomness, no physics chaos, and no broad player improvisation. Each puzzle is refined to deliver one specific logical idea.

That contrast is the larger lesson. There is no one perfect way to design puzzles. Some great games are crafted by a master designer who takes the player through a precise sequence of ideas. Other great games use randomness, physics, and open-ended tools to make the player feel like the master craftsperson.

Mosa Lina succeeds because it commits to the second path. It does not hide its randomness or pretend every room has a perfect authored solution. It embraces the instability and asks the player to create meaning inside it.

That is the joy of games: there is room for every type. And if a designer does not like what the current landscape offers, they can make something different and show how another kind of play might work.