🖥️ PC2011MojangHARD

Minecraft rebuild deck

HARD

You will get the loop working. Then the real work starts.

The biggest surprise on this list is how small the core actually is — noise terrain, a raycast, a recipe table, a light propagator and a mesher is most of it, and a competent agent writes all five. Where it fails you is the mesher: you will get a cube per block, it will run beautifully in a 32-cube test world, and it will die the moment you stream chunks. Demand the quad count on screen from the first commit and refuse the build until it is in the hundreds. What no prompt reaches is the fifteen years of things that got added because players were already doing them.

Difficulty: HARD · 3 of 5
First playable: A project
The original: 10 team-years
Released: 2011 · PC
Genre: Voxel Sandbox

The one-shot build prompt

Build the core of an infinite voxel sandbox: one biome set, survival loop only. No alternate dimensions, no circuitry, no multiplayer. three.js, chunked world, browser.

Requirements:
- Chunk meshing is the project. Store 16x16x256 chunks as a flat Uint8Array and build one merged BufferGeometry per chunk, culling faces against neighbouring chunks and greedy-merging coplanar same-block faces into larger quads. Never one mesh per block. On an edit, rebuild that chunk plus the neighbour if the edit touched a boundary. Build this first and print the quad count on screen.
- Terrain from layered noise off a single seed: a height field, a second field selecting biome, caves carved by thresholded 3D noise, ore in veins by depth band.
- Break and place with a voxel DDA raycast returning the hit voxel and its face normal, so placement lands on the correct side. Break time scales with block hardness and held tool.
- Lighting as a queue-based flood fill: skylight at full strength, torch light decrementing one per block, re-propagated locally on edit rather than recomputed.
- Survival: health, hunger draining with activity, a 3x3 grid driven by shaped and shapeless recipes in JSON, a nine-slot hotbar, a twenty-minute day cycle, hostile mobs spawning below a light threshold.
- Persist modified chunks only, as a diff against the seed, in IndexedDB.

Definition of done: I walk a thousand blocks with chunks streaming in, dig into a cave, place a torch, and the light spreads round the corner at 60fps.

Before you start🧰 The toolkit — models, MCP servers, art and 3D

What you lose

Fifteen years of accretion — the Nether, the End, enchanting, villagers, boats, bees — almost none of which is the core loop and all of which is why people stay
Redstone, a Turing-complete circuit language nobody sat down and designed, which half the community treats as the actual game
Servers holding hundreds of players with authoritative block edits and chunk streaming that does not fall over
A terrain generator rewritten repeatedly over a decade so that walking in one direction keeps paying out

What to steal

Greedy meshing — merge coplanar same-block faces so a chunk of a million blocks draws as a few hundred quads.

Draw calls and vertex count are what kill grid-based renderers, not the number of cells. The same sweep collapses tilemap layers, collision geometry and heightmaps, and it is about thirty lines once you have seen it done.

ScopeA weekend— build just this, not the game

Build a greedy meshing demonstrator as a single self-contained HTML file, Canvas 2D, no build step.

Requirements:
- A 128x128 grid of tile ids generated from noise into a few large contiguous regions plus scattered single tiles.
- Naive pass: one rectangle per tile. Count them.
- Greedy pass: sweep row by row; for each unvisited tile extend a run right while the id matches, then extend that run downward while every tile in the next row matches and is unvisited, marking cells consumed as you go. Emit one rectangle per merged region. Count them.
- Draw both passes with a one-pixel outline per rectangle so the merge is obvious, and a key to toggle between them.
- Show both counts and the ratio in the corner. Regenerate on a keypress.
- Comment the line where a 3D chunk would run this same sweep once per axis per slice.

Definition of done: the greedy pass draws a visually identical grid with an order of magnitude fewer rectangles, and the counter proves it.

Make it different

Not a clone — Minecraft re-themed and cut down to something you can finish. The angle is the important reel: it is what turns a project into a weekend. 24,360 ways to land.

On Demand Hands Off

Minecraft, but an auto-battler — you set up, it plays itself, set in gig economy streets. Rendered in comic-book halftone.

Themegig economy streets

Anglean auto-battler — you set up, it plays itself

Lookcomic-book halftone

Build a browser game called "On Demand Hands Off".

The pitch in one line: a voxel sandbox in the spirit of Minecraft, but an auto-battler — you set up, it plays itself, set in gig economy streets.

WHAT THIS IS
Take the SHAPE of Minecraft — its core loop, and whatever makes that loop good — and rebuild it as your own game in a different world at a smaller scope. Use your own names, art and audio, which you want anyway: a reskin of somebody else's game is less interesting than a new one that works the same way.

THE ANGLE — this is the important constraint, honour it above everything else
An auto-battler — you set up, it plays itself. The player only makes decisions between rounds. Combat resolves deterministically from board state and seed, so a replay is exact.
This is what makes the project finishable. If a decision would push the scope back towards the original's, take the smaller option every time.

THE THEME
Gig economy streets. Ratings, tips and routing under a harsh clock. The city is a delivery graph and the player is the last mile.
The theme is not a coat of paint. Let it change what the mechanics mean — a reload, a health pack and a locked door should all be things that make sense in this world and nowhere else.

LOOK
Comic-book halftone. Bold ink, Ben-Day dots, speed lines on hits, and panel-border flashes on big moments.

REQUIREMENTS
- Runs in a browser with no build step. Canvas 2D unless the angle genuinely requires 3D, in which case three.js.
- Fixed-timestep update loop with interpolated rendering, so it behaves the same at 60Hz and 144Hz.
- Every tuning constant in one CONFIG object at the top, exposed as live sliders in a debug panel. You will find the feel by dragging those, not by prompting.
- All content — levels, entities, balance numbers — in JSON or plain data files, never inline in the logic.
- Audio generated with the Web Audio API rather than asset files.
- Respect prefers-reduced-motion: keep the fades, drop the shake and the parallax.

BUILD ORDER
1. The core loop with placeholder rectangles, at the reduced scope the angle demands. No theme, no art, no audio. Make it fun as rectangles first.
2. The angle's consequences. Cutting to bots, or to turn-based, or to one level changes the design rather than just shrinking it — find out how and rewrite step 1.
3. The theme, as content and data.
4. The look, applied last as a rendering layer over a game that already works.

Start by writing a one-page plan: the core loop, what the angle removes and what that frees you to do properly, and what you are deliberately leaving out. Wait for me to approve it before writing code.

Prior art — read before you build

Half the time one of these already does what you wanted, and the other half it saves you a week of solving a problem someone documented in 1997.

Did you build it?

0Share on X

One count per game. No account — it is tied to a salted hash of your IP, which is not stored in a form anyone can reverse. Every build adds 10 team-years to the counter on the homepage.

Then try these

Or roll something nobody has built yet →

Questions

How hard is Minecraft to rebuild with an AI agent?

Yes — every game is buildable at some scope. The question is how hard, and Minecraft is HARD (3 of 5). You will get the loop working. Then the real work starts. The core is five small systems and an agent writes all five; the greedy mesher decides whether you have a world or a slideshow by chunk two hundred, and past that Minecraft is fifteen years of accretion you would have to invent yourself. Reckon on a project to something you can actually play — not to something that matches the 10 team-years Mojang spent.

What makes Minecraft hard to rebuild?

The core is five small systems and an agent writes all five; the greedy mesher decides whether you have a world or a slideshow by chunk two hundred, and past that Minecraft is fifteen years of accretion you would have to invent yourself. That is why it sits at HARD rather than a tier either side of it. If you want the idea without the project, the transferable part is greedy meshing — merge coplanar same-block faces so a chunk of a million blocks draws as a few hundred quads. — a weekend of work.

How long does it take to build a Minecraft clone?

Wrong question, and it is worth saying why. Nobody rebuilds Minecraft — the original took Mojang on the order of 10 team-years, released in 2011 on PC. What you can do is reach something playable and genuinely yours in about a project, and then keep going for as long as it stays interesting. Any figure that claims otherwise is measuring a prototype and calling it a game.

What do I lose by building my own Minecraft?

Fifteen years of accretion — the Nether, the End, enchanting, villagers, boats, bees — almost none of which is the core loop and all of which is why people stay. Redstone, a Turing-complete circuit language nobody sat down and designed, which half the community treats as the actual game. Servers holding hundreds of players with authoritative block edits and chunk streaming that does not fall over. Those are the parts a prompt will not hand you.

Which AI coding agent should I use to build it?

Any of Claude Code, Codex or Cursor will handle this prompt. Paste it as your first message, let the agent scaffold the project, then iterate on feel — the second, third and tenth prompts are where a game actually gets good. Work in an empty folder or a fresh git branch so you can throw it away cheaply.

Get the next game

One email when new games go on the list. No other reason to email you, ever.