🎮 PS32009Naughty DogVERY HARD

Uncharted 2: Among Thieves rebuild deck

VERY HARD

You will get a good prototype. The product is what comes after.

The local-space decision is the whole build. Make it in the first prompt and everything downstream is ordinary platformer work; make it in week two and you are rewriting the controller, the collision and the camera together. This stays a NO, and it is the honest NO on the list, because the product here is not a system — it is timing. The carriage that detaches on the exact beat you happened to look at it was placed there by a director, frame by frame, and it happens once. There is no prompt for a director.

Difficulty: VERY HARD · 4 of 5
First playable: A project
The original: 200 team-years
Released: 2009 · PS3
Genre: Action Adventure

The one-shot build prompt

Build ONE moving set piece: a climbing traversal along the roof of a train travelling at speed. three.js, a chain of carriages, no shooting, no story.

Requirements:
- The load-bearing decision, made before the first line of code: simulate the entire scene in the train's local space. The train is a root transform moving through the world; the player, the props, the physics and the camera basis are all parented to it, and you convert to world coordinates only at render time. Jump arcs, gravity and collision resolve locally. Simulate the player in world space and add the train's velocity back each frame instead, and you will spend the whole build fighting the player sliding off the rear of the carriage. Retrofitting this later means rewriting the controller, the collision and the camera together.
- Ledge traversal: hang, shimmy left and right, jump between handholds, mantle. Handholds are authored edge segments with a facing normal, and the reachable set from any hang is a scored search over nearby segments — input direction first, distance second, line of sight third.
- Crumbling handholds that detach 1.2 seconds after taking weight, so standing still is fatal.
- Wind and carriage sway: a low-frequency roll applied to the root, which the local simulation gets for free.
- Parallax: pass the train's world velocity to the skybox and the ground plane only, so the carriage feels still underfoot while the world tears past.

Definition of done: I jump a gap between two carriages at speed, and the train never once shoves me backwards.

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

What you lose

The set pieces, scripted beat by beat — the collapsing building is one long piece of bespoke timing that exists exactly once
Performance capture and a written script, which is what makes climbing a wall feel like a scene rather than an exercise
Traversal level design where every handhold is placed to read as a route from thirty metres away
Encounter pacing that pushes you out of cover on a rhythm you never consciously notice

What to steal

Grab magnetism — a jump that ends near a ledge is scored onto it, and the hand animation is warped so the grab lands exactly on the edge.

The player gets credited for an input they did not quite make, and the animation lies convincingly enough that they never notice being helped. It is the whole difference between a climbing game that feels generous and one that feels broken. Transfers to any game with a landing: a platformer's edge, a cover snap, a docking manoeuvre, a melee lunge.

ScopeA weekend— build just this, not the game

Build a single-file HTML page, Canvas 2D, side-on, one stick figure and four ledges, demonstrating grab magnetism.

Requirements:
- Ledges are edge segments with a position and a facing direction. The figure jumps on an ordinary ballistic arc under gravity.
- Each frame of the jump, score every ledge within 0.6 body-heights of the projected landing point: alignment with the held input direction first, distance second, rejecting anything behind a wall. If one scores above a threshold, commit.
- Once committed, do not teleport. Blend the body's position onto the ledge across the final 0.15 seconds of the arc with an ease-out, and separately drive the hand point to the exact edge coordinate so contact is pixel-accurate. The body follows the hand, not the reverse.
- A debug toggle drawing the tolerance radius, every scored candidate, and the raw uncorrected arc.
- A second toggle that disables magnetism entirely.

Definition of done: with magnetism off I miss the same jump repeatedly, and with it on I do not notice I am being helped until I turn the debug draw on.

Make it different

Not a clone — Uncharted 2: Among Thieves 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. 5,880 ways to land.

Leviathan Turn Order

Uncharted 2: Among Thieves, but turn-based instead of real-time, set in a kaiju attack. Rendered in CRT arcade with scanlines.

Themea kaiju attack

Angleturn-based instead of real-time

LookCRT arcade with scanlines

Build a browser game called "Leviathan Turn Order".

The pitch in one line: a action adventure in the spirit of Uncharted 2: Among Thieves, but turn-based instead of real-time, set in a kaiju attack.

WHAT THIS IS
Take the SHAPE of Uncharted 2: Among Thieves — 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
Turn-based instead of real-time. Time only advances when the player acts. Removes every timing, netcode and animation-blending problem and turns reflexes into decisions.
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
A kaiju attack. Scale as the entire design: the player is either the enormous thing or a very small thing near it, and the city is a physics object.
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
CRT arcade with scanlines. Post-process pass with scanlines, barrel distortion, phosphor bloom and slight chromatic aberration at the edges. Keep it subtle enough to read text through.

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 200 team-years to the counter on the homepage.

Then try these

Or roll something nobody has built yet →

Questions

How hard is Uncharted 2: Among Thieves to rebuild with an AI agent?

Yes — every game is buildable at some scope. The question is how hard, and Uncharted 2: Among Thieves is VERY HARD (4 of 5). You will get a good prototype. The product is what comes after. Simulate the scene in the train's local space and the traversal becomes ordinary platformer work; what you cannot prompt is a director placing a collapse on the exact frame you happened to be looking at it. Reckon on a project to something you can actually play — not to something that matches the 200 team-years Naughty Dog spent.

What makes Uncharted 2: Among Thieves hard to rebuild?

Simulate the scene in the train's local space and the traversal becomes ordinary platformer work; what you cannot prompt is a director placing a collapse on the exact frame you happened to be looking at it. That is why it sits at VERY HARD rather than a tier either side of it. If you want the idea without the project, the transferable part is grab magnetism — a jump that ends near a ledge is scored onto it, and the hand animation is warped so the grab lands exactly on the edge. — a weekend of work.

How long does it take to build a Uncharted 2: Among Thieves clone?

Wrong question, and it is worth saying why. Nobody rebuilds Uncharted 2: Among Thieves — the original took Naughty Dog on the order of 200 team-years, released in 2009 on PS3. 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 Uncharted 2: Among Thieves?

The set pieces, scripted beat by beat — the collapsing building is one long piece of bespoke timing that exists exactly once. Performance capture and a written script, which is what makes climbing a wall feel like a scene rather than an exercise. Traversal level design where every handhold is placed to read as a route from thirty metres away. 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.