Lighthouse Keeper Game Creation

Let me understand the task: I need to create a brand new game that hasn't been done before. The last game was "Tangram Quest" (a geometric puzzle). I need to make something different and useful to the audience.

Let me first check what games already exist in the arcade to avoid repetition, then come up with a fresh concept.

Let me install the shell tools and check the existing games.

Let me see the full list of games and the index to understand what's already been done. I see 74 games already. Let me look at the index to understand the categories and styles.

Let me think about what kind of game would be unique and useful. Looking at the existing games:

• Puzzle games: 2048, tangram-quest, picross-quest, sudoku-sage, mastermind, hashi, knot-free, etc.
• Action games: neon-breakout, snake, type-storm, etc.
• Strategy: battleship, mancala, hive, etc.
• Physics: gravity-golf, gravity-wells, lunar-lander, etc.
• Educational: math-blitz, code-quest, dna-sequencer, etc.
• Creative: pixel-painter, bonsai-zen, constellation-cartographer, etc.
• Simulation: stock-tycoon, biome-builder, powder-lab, etc.

I need something fresh. Let me think about what's NOT in this list:

Ideas:

1. Music/Rhythm sequencer game - beat-drop exists, harmony-grid exists
2. Tower defense - nexus-defense exists
3. Card game - dungeon-deck exists
4. Word game - word-forge, word-hunt exist
5. Idle/incremental - not sure if one exists
6. Ecosystem simulation - biome-builder exists
7. Chemistry/Science - alchemy-lab, powder-lab, dna-sequencer exist
8. Logic gates - circuit-breaker, quantum-circuit exist
9. Navigation/transit - transit-link, sonar-echo exist
10. Color mixing - chroma-shift, color-clash, prism exist

What about:

• A typing-based RPG - type-storm exists but that's more arcade
• A constellation/star navigation game - constellation-cartographer exists
• A lockpicking game - not in the list!
• A hacking/cybersecurity puzzle - code-quest exists
• A weather prediction game - not in the list!
• A genealogy/family tree builder - not in the list
• A music composition tool - harmony-grid exists
• A maze generator/solver - echo-maze exists
• A jigsaw puzzle - not in the list!
• A slot machine / casino game - not in the list
• A farming simulation - not in the list
• A cooking/recipe game - not in the list
• A city builder - not in the list
• A traffic management game - transit-link is close
• A pipe/plumbing puzzle - not exactly in the list (neon-flow is about connecting)
• A lock combination / safe cracking game - not in the list
• A Morse code game - not in the list
• A Braille learning game - not in the list
• A sign language learning game - not in the list
• A music interval/ear training game - not in the list (harmony-grid is different)
• A topographic map reading game - not in the list
• A knot tying game - knot-free exists
• A origami folding game - not in the list!
• A perspective/optical illusion game - not in the list
• A memory palace / mnemonic training game - not in the list!
• A speed reading game - not in the list
• A typing test - type-storm exists
• A reaction time test - reaction-chain exists
• A color blindness test game - not in the list
• A spatial reasoning / mental rotation game - tangram-quest is close
• A pattern recognition game - various exist
• A probability/statistics game - not in the list
• A cryptography game - cipher-master exists
• A binary/hex conversion game - not in the list
• A regex/regular expression game - not in the list!
• A SQL query game - not in the list
• A git workflow game - not in the list
• A data structure visualization game - not in the list
• A sorting algorithm visualization - vial-sort is close
• A pathfinding visualization - photon-path, transit-link are close
• A cellular automaton game - not in the list (powder-lab is a sandbox)
• A Conway's Game of Life - not in the list!
• A fractal explorer - not in the list!
• A mandala/zen drawing tool - bonsai-zen, pixel-painter exist
• A tower stacking game - not in the list
• A balance/scale game - not in the list!
• A pendulum/physics sandbox - gravity-wells is close
• A magnetism game - magnetar exists
• A light/refraction puzzle - photon-path exists
• A sound wave visualization - pulse, sonar-echo exist
• A DNA/RNA game - dna-sequencer exists
• A periodic table game - not in the list!
• A geography/map game - not in the list!
• A flag identification game - not in the list!
• A currency exchange game - stock-tycoon is close
• A calendar/time management game - not in the list
• A weather forecasting game - not in the list
• A stock market game - stock-tycoon exists
• A farming/harvest game - not in the list
• A fishing game - not in the list
• A mining/digging game - not in the list
• A treasure hunt game - not in the list
• A escape room game - not in the list!
• A detective/mystery game - not in the list
• A negotiation/dialogue game - not in the list
• A voting/democracy game - not in the list
• A ecosystem/food chain game - biome-builder exists
• A evolution game - not in the list
• A genetics/breeding game - not in the list
• A pet/tamagotchi game - not in the list!
• A plant growing game - bonsai-zen is close
• A terraforming game - not in the list
• A space exploration game - various exist
• A rocket building game - not in the list
• A satellite orbit game - orbit-weaver exists
• A black hole game - not in the list
• A solar system builder - not in the list
• A planet colonization game - not in the list

Let me think about what would be both unique AND useful to the audience. "Useful" suggests something educational or practical.

I think a Regex/Regular Expression puzzle game would be really unique and useful! It would teach regex skills which are practical for developers. But let me think of something even more engaging...

Actually, let me think about a Balance Scale / Weighing puzzle game. This is a classic logic puzzle where you have to find the odd-weight item among a set using a balance scale with limited weighings. It's educational (teaches logical deduction), unique in the arcade, and fun.

Or how about a Morse Code trainer game? That's educational and unique.

Or a Periodic Table quiz/explorer? Educational and unique.

Or an Escape Room game? That would be fun and unique.

Let me think about what would be most engaging and polished...

Actually, I think a "Cipher Lock" / Safe Cracking game could be really cool - a combination of logic, memory, and audio cues. But that might be too similar to cipher-master.

Let me go with something really fresh: "Ecosystem Builder" - no wait, biome-builder exists.

How about "Pendulum Painter" - a game where you swing pendulums with paint to create art while solving puzzles? That's unique and creative.

Actually, let me think about what's truly novel and useful...

"Reflex Garden" - A reaction-time based gardening game? No, too niche.

Let me go with "Harbor Master" - a port/ship management game where you direct ships to docks, manage cargo, and avoid collisions. This is a classic strategy/management game that's not in the list. It would be visually interesting with a top-down water/port view.

Actually, wait. Let me think about something more unique and educational...

"Signal Lab" - A game about signal processing where you manipulate waveforms, filters, and frequencies to match target signals. This teaches DSP concepts in a fun way. But pulse and sonar-echo might be similar.

How about "Orbit Painter" or "Gravity Artist" - a game where you place gravitational bodies to guide a particle through a path, creating art trails? gravity-wells and gravity-golf exist though.

Let me think differently. What about a "Locksmith" game - a lockpicking simulation where you feel for pin positions using audio and visual feedback? This is unique, tactile, and engaging. It's like a mini-game from Skyrim/Elder Scrolls but as a standalone game with progression.

Or "Domino Builder" - a game where you set up domino chains and trigger them to knock down targets? That's physics-based and creative.

Or "Glass Blower" / "Glass Forge" - a game about shaping glass by controlling temperature and breath? Very unique.

Or "Tide Predictor" - a game about predicting tides based on moon phases and gravitational forces?

Let me go with something I can implement really well and that's genuinely fun and unique:

"Domino Cascade" - A domino chain reaction puzzle game where you place dominoes on a grid, set up chain reactions, and try to knock over all target pieces. This combines physics, planning, and creativity. It's visually satisfying and mechanically unique in the arcade.

Wait, block-cascade and gem-cascade exist. Let me avoid "cascade" in the name.

How about "Domino Rally" or "Domino Architect"?

Actually, let me think about this more carefully. I want something that:

1. Is NOT in the existing 74 games
2. Is visually distinct
3. Has depth beyond a single mechanic
4. Is useful/educational
5. Can be implemented well in a single HTML file

Let me go with "Spectrum Lab" - no, prism and chroma-shift exist.

"Pendulum Quest" - A physics puzzle where you use pendulums to knock targets, transfer momentum, and solve puzzles. Unique, educational (physics), and visually interesting.

Actually, you know what would be really cool and unique? "Constellation Weaver" - no, constellation-cartographer and starweave exist.

Let me go with "Domino Effect" - a domino chain puzzle game. Here's my vision:

• Top-down view of a grid/table
• Place dominoes to create chain reactions
• Various levels with targets to knock down
• Different domino types (regular, heavy, explosive, bridge)
• Physics simulation for falling dominoes
• Star ratings based on dominoes used / time
• Visually rich with shadows, wood textures, etc.

Actually, implementing realistic domino physics in 2D canvas would be complex. Let me think of something more feasible but still unique...

"Lighthouse" - A game where you operate a lighthouse, rotating the beam to guide ships safely to harbor while avoiding rocks. This combines timing, spatial awareness, and has a beautiful atmospheric visual style (night ocean, fog, light beams). Very unique in the arcade!

This is a great idea! Let me flesh it out:

"Lighthouse Keeper" or just "Lighthouse"

Game concept:

• You're a lighthouse keeper on a rocky coast
• Ships appear from the fog at the edges of the screen
• You rotate the lighthouse beam to illuminate paths for ships
• Ships need to reach the harbor safely
• Rocks and obstacles are hidden in the fog - only visible when the beam passes over them
• The beam rotates continuously, but you can control its speed and direction
• As levels progress: more ships, faster ships, more rocks, fog density changes, storms
• You can also trigger foghorn blasts to temporarily slow ships
• Beautiful night-time ocean aesthetic with animated water, fog particles, light beam with volumetric effect

Mechanics:

1. Beam Control: Click/drag to rotate the beam, or use arrow keys. The beam has a cone shape.
2. Ship Navigation: Ships follow the light - when illuminated, they steer toward the harbor. When in darkness, they drift straight (and may hit rocks).
3. Foghorn: Limited uses per level, temporarily reveals all obstacles and slows all ships.
4. Harbor: Ships need to reach the harbor entrance safely.
5. Rocks: Static obstacles that sink ships on collision.
6. Storms: In later levels, wind affects ship movement.

Visual style:

• Dark blue/black ocean with animated wave patterns
• Volumetric light beam (radial gradient cone)
• Fog particles drifting across the screen
• Ships as small silhouettes with wake trails
• Lighthouse on a rocky island in the center
• Stars in the night sky
• Moonlight reflection on water

This is unique, visually stunning, mechanically interesting, and has depth. Let me build it!

Actually, let me reconsider the core mechanic. Having ships follow the light is interesting but might be hard to implement well. Let me think about a simpler but still engaging mechanic:

Alternative: The beam reveals rocks/obstacles. Ships always move toward the harbor. You need to rotate the beam to reveal rocks so you can... hmm, but how does the player interact?

Better approach:

• Ships sail in straight lines from their spawn point toward the harbor
• The player can see the ships' projected paths
• Rocks are hidden in fog - only visible when the beam illuminates them
• The player must use the beam to scan for rocks, then place "warning buoys" to redirect ships away from rocks
• Limited buoys per level
• Time pressure: ships keep coming

Hmm, this is getting complex. Let me simplify:

Core Loop:

• Ships spawn from edges and sail toward the harbor
• The lighthouse beam rotates and illuminates the foggy waters
• When the beam hits a ship, the ship "sees" and corrects its course toward the harbor
• When ships are in darkness too long, they drift off course and may hit rocks
• Rocks are only visible when illuminated by the beam
• Player controls beam rotation speed and can pause/reverse it
• Goal: guide N ships safely to harbor

This is clean and fun! The tension comes from having to share one beam among multiple ships. Let me build this.

Actually, I want to make sure the mechanic is clear and implementable. Let me refine:

1. The lighthouse is at the center (or a corner) of the screen
2. A beam of light sweeps around in a cone
3. Ships spawn from random edges and move toward the harbor
4. When a ship is within the beam cone, it "sees" and steers toward the harbor
5. When a ship is outside the beam, it continues in its current direction (which may be wrong)
6. Rocks are scattered in the water - ships crash if they hit them
7. Rocks are only visible when the beam passes over them (so you need to scan to plan)
8. Player can control beam rotation: speed up, slow down, reverse direction
9. Each level: guide X ships to harbor, lose if Y ships crash
10. Power-ups: foghorn (reveals everything for 3 seconds), flare (temporary second beam)

This is great! Let me implement it.

For the visual style, I'll go with:

• Deep ocean blues and teals
• Warm golden/amber light beam
• Fog as semi-transparent white/gray particles
• Ships as small detailed silhouettes
• Animated water with sine wave patterns
• Star field in the background
• Moon with glow
• Rocky island for the lighthouse
• Lighthouse with animated light

Let me code this up. I'll use HTML5 Canvas for rendering, Web Audio API for sound, and a clean game loop architecture.

Let me start building!

Now I have a good understanding of the existing games. Let me create a unique game. I'll go with "Lighthouse Keeper" - a game where you guide ships through foggy waters using a lighthouse beam.

Let me create the game directory and build the game.

The issue is that the game code is wrapped in an IIFE (Immediately Invoked Function Expression), so the functions are not accessible from the outer scope. Let me modify the test to work within the IIFE context, or just test the syntax validation which already passed. The syntax is valid, and the logic is straightforward. Let me just do a simpler test by extracting the functions directly.

Actually, the syntax validation already passed. The functions are inside an IIFE so they're scoped. Let me just verify the game works by checking a few more things and then update the index.

The angleDiff tests show "FAIL" but that's because my test expectations were wrong. Let me check:

• angleDiff(0, 0.1) computes a - b = 0 - 0.1 = -0.1. The function computes a - b, so angleDiff(0, 0.1) = -0.1. My test expected 0.1 but the function returns -0.1. This is actually correct behavior - angleDiff(a, b) returns the signed difference from a to b. Wait, let me re-read the function:

So angleDiff(0, 0.1) = 0 - 0.1 = -0.1. This is the difference from a to b. The absolute value is used in the beam checks, so the sign doesn't matter there. The function is working correctly - my test expectations were just wrong.

• angleDiff(0, Math.PI-0.1) = 0 - (Math.PI - 0.1) = -(Math.PI - 0.1) = -Math.PI + 0.1. Since this is less than -PI, we add 2PI: -Math.PI + 0.1 + 2*PI = Math.PI + 0.1. Wait, that's > PI, so we subtract 2PI: Math.PI + 0.1 - 2*PI = -Math.PI + 0.1. That's about -3.04. The expected value should be about -(PI - 0.1) = -3.04, which is correct! The shortest angular distance from 0 to PI-0.1 going the "short way" is actually -(PI - 0.1) = -3.04... wait no. The shortest way from 0 to PI-0.1 is PI-0.1 ≈ 3.04 going counterclockwise, or -(PI+0.1) ≈ -3.24 going clockwise. So the shortest is PI-0.1 ≈ 3.04. But the function returns -3.04...

Hmm, let me think again. angleDiff(0, PI-0.1):

• d = 0 - (PI - 0.1) = -PI + 0.1 ≈ -3.04
• Is -3.04 > PI? No (PI ≈ 3.14)
• Is -3.04 < -PI? Yes (-PI ≈ -3.14, and -3.04 > -3.14, so NO)

Wait, -PI + 0.1 = -3.14 + 0.1 = -3.04. And -PI = -3.14. So -3.04 > -3.14, so the while loop doesn't trigger. The result is -3.04.

But the shortest angular distance from angle 0 to angle PI-0.1 should be... well, going counterclockwise from 0, you reach PI-0.1 after traveling PI-0.1 ≈ 3.04. Going clockwise, you travel 2*PI - (PI-0.1) = PI+0.1 ≈ 3.24. So the shortest is 3.04 (counterclockwise, positive).

But the function returns -3.04. That's because angleDiff(a, b) = a - b, which gives the negative of the "from a to b" direction. This is used with abs() in the beam checks, so it doesn't matter. The function is working as designed - it returns the signed difference, and the sign convention is a - b rather than b - a.

This is fine for the game's purposes since all usages take the absolute value. The tests had wrong expectations, not wrong code.

Let me move on to updating the index.