BUILDING A GAME IN 7 DAYS

TL;DR

The seven-day build targets a complete, playable roguelite tower defense using Lua and Cursor, with most development happening live and daily devlogs planned.

Briefing Cornell Notes

Briefing

A seven-day, from-scratch tower defense game is being built with Lua and Cursor, but the pitch isn’t just “ship fast”—it’s a specific design blend: tower-defense core mechanics paired with deckbuilding and roguelite-style progression. The goal is to finish a playable game by the end of the week while live-streaming the bulk of the development, with daily devlogs planned so progress is visible without watching the entire stream.

At the center is a card-driven tower placement system. Players start each run with a deck and a draw hand, then place towers on a tower-defense map made of multiple mazes/waves (the team explicitly wants a familiar, non-novel pathing approach rather than something like a complex “mazing” system). Towers are represented by cards: when a tower card is played, it’s exhausted into an “exhaust pile,” meaning it can’t be used again for the rest of the map. As enemies move through waves, towers deal damage and can upgrade when they perform well enough—upgrades arrive through a Vampire Survivors-like stat screen with multiple percentage-based modifiers (examples mentioned include damage, fire rate, and critical chance), but the upgrades are randomized so players can’t always optimize for a single stat every time.

Between maps, the game shifts into a shop-and-mutation loop. After completing a map’s waves, players get a shop/upgrade screen with tower-specific upgrades that can directly modify the tower card they already used, plus additional cards that are added to the deck. These new cards enable mid-run “mutations” of towers based on what’s in the player’s hand—potentially lasting for a round, a wave, or longer. The design aims for conditional synergies (“if crit then do X,” “attack faster,” or apply poison/splash effects), and it leans into tactical tradeoffs against enemy modifiers such as double-speed foes.

Tower upgrades are also meant to branch into distinct classes. A sniper example illustrates the philosophy: converting a tower into a higher-damage variant can come with a sacrifice to shooting speed. Another concept is intentionally debuffing a tower’s firing rate to unlock a larger damage multiplier the longer it doesn’t shoot, pushing players away from the “always maximize fire rate” mindset. The shop then adds global modifiers—things like base damage or fire-rate boosts for all arrow towers or all cannons—plus longer-lasting equipment-style effects (rings/amulets were floated as possibilities).

Boss fights and relic-style choices bring in the Slay the Spire influence. After defeating bosses, players gain relic options that can grant more actions per turn or other modifiers, with some relics potentially hiding information about upcoming enemy types to force imperfect planning. The run is designed to end in three broad outcomes: a crushing loss, a narrow win, or a dominant win—plus multiple ways to succeed, including non-lethal strategies like heavy shielding that lets the player “tank” enemies. The team also hints at late-game aura/level multipliers that apply across multiple slots for limited durations, reinforcing the roguelite feel.

The week’s structure—live building in a former water tower turned Airbnb/game studio, multiple characters planned (a “everyday hero,” a wizard class, and a futurist), and a Twitch-focused completion target—serves the same purpose as the mechanics: make the game fun, varied, and not another monetized mobile tower defense.

Cornell Notes

The project aims to build a roguelite tower defense in seven days using Lua and Cursor, mixing tower placement with deckbuilding and randomized progression. Towers are played as cards from a draw hand and then exhausted for the rest of the map, while performance triggers randomized stat upgrades (e.g., damage, fire rate, crit). Between maps, players use a shop to apply tower-specific upgrades and add new cards that can mutate towers mid-run based on what’s in hand, with conditional synergies and tradeoffs. Bosses grant relic-like choices inspired by Slay the Spire, sometimes with hidden information about upcoming enemies. The design targets multiple win conditions (including shielding/tanking) and three broad end states: lose badly, win narrowly, or win big.

How does the game turn tower defense gameplay into a deckbuilding loop?

Towers are represented by cards in a player’s draw hand. When a player selects a tower card (e.g., an archer or cannon), it’s placed on the map and then exhausted into an “exhaust pile,” so it can’t be played again for the rest of that map. Enemies advance in waves; towers damage them and can upgrade when they perform well enough. After the map’s waves, the player enters a shop/upgrade phase to buy tower-specific upgrades and add new cards to the deck, which then influence future waves and maps.

What role does randomness play, and why isn’t it just “always pick the best stat”?

Upgrades after tower performance come with a stat-selection screen similar to Vampire Survivors, with percentage modifiers such as 15% more damage, 15% more fire rate, and 10% more critical chance. The team emphasizes that the upgrade options should be randomized so players can’t reliably force the same optimization every time. That randomness is paired with roguelite learning: players must adapt their strategy to the upgrades and cards they actually receive.

How do mid-run card additions change towers after they’ve already been placed?

The shop can grant two kinds of tower-related rewards: (1) tower-specific upgrades that apply directly to the tower card the player used, and (2) new cards added to the deck that can mutate towers mid-game. These mutations can be conditional (examples include “if crit then send a lightning bolt” or “if crit then attack faster”) and can last for a round, a wave, or longer, depending on the card. This creates a balancing problem between tower stats, the deck’s mutation effects, and the enemy mix.

What kinds of tradeoffs are intended to prevent “one god tower” strategies?

The design explicitly targets branching tower classes and tradeoffs. A sniper-class example converts a tower into something with much higher damage but reduced shooting speed. Another concept is debuffing shooting speed on purpose to gain a larger damage multiplier the longer the tower doesn’t fire. These mechanics aim to make different builds viable rather than a single dominant fire-rate stacking approach.

How do bosses and relics affect planning and run outcomes?

Bosses drop relic-style choices inspired by Slay the Spire. Relics can grant more actions per turn or other modifiers, and some may hide information—such as not revealing what enemy types are coming next—forcing players to guess and adapt. The run is designed to end in three broad outcomes: a crushing loss, a narrow win, or a dominant win, with multiple paths to victory including shielding/tanking rather than only killing enemies.

What global progression layers are planned beyond per-tower upgrades?

After tower and deck mutations, the shop includes global modifiers that can apply across tower categories (e.g., base damage increases for all arrow towers or fire-rate boosts for cannons). Equipment-like effects such as rings or amulets were suggested as persistent modifiers. Late in the run, level multipliers and aura effects are hinted at, including limited-duration slots and aura effects lasting for multiple games, adding another strategic layer on top of tower-level changes.

Review Questions

Describe the lifecycle of a tower card from draw hand to exhaust pile. What does that imply for how players build during a map?
Give one example of a tower-class tradeoff (like sniper) and explain how it changes player decision-making compared with maximizing fire rate.
How do relics from bosses introduce uncertainty, and how does that uncertainty connect to the game’s three end-state outcomes?

Key Points

1
The seven-day build targets a complete, playable roguelite tower defense using Lua and Cursor, with most development happening live and daily devlogs planned.
2
Towers are card-based: playing a tower card places it and exhausts it for the rest of the map, forcing repeated strategic choices across waves.
3
Tower upgrades trigger after performance and use randomized stat selections (e.g., damage, fire rate, critical chance) to prevent one guaranteed optimization path.
4
Between maps, a shop provides both tower-specific upgrades and new deck cards that can mutate towers mid-run through conditional effects and timed durations.
5
Tower progression is designed to branch into classes with meaningful tradeoffs (e.g., sniper: higher damage but slower shooting).
6
Bosses grant relic-like choices inspired by Slay the Spire, sometimes hiding upcoming enemy information to increase adaptation pressure.
7
Victory isn’t limited to killing: heavy shielding and tanking are intended as alternative win conditions, supporting multiple build archetypes.

Highlights

Tower placement is driven by cards: once a tower card is played, it goes to an exhaust pile and can’t be reused for the rest of the map.

Upgrades arrive through randomized stat screens, combining roguelite unpredictability with tactical adaptation rather than fixed “best build” paths.

Sniper-class conversion is framed as a deliberate tradeoff—more damage at the cost of shooting speed—designed to break the fire-rate meta.

Boss relic choices may hide enemy-type information, turning planning into a guessing game while still aiming for narrow or dominant wins.

Topics

Tower Defense
Deckbuilding
Roguelite Progression
Lua Development
Cursor Workflow