Village Buildings

The bell tower periodically offers the player a village building to construct. Each building permanently boosts one of the three village "support" scores. This page lists every building in buildings.json with its exact per-level material cost and per-level bonus, and explains the village.building_milestones schedule that decides how often the bell offers one. It also covers the class landmarks — the champion-only signature buildings that sit outside the bell schedule.

In plain terms — what a building does

The village runs on three support scores: food, shelter, and safety. Each building feeds exactly one of them. Building it, and leveling it up, adds a flat number to that metric's score.

Two words there need defining. Support is an internal points score; the village's overall health is set by whichever of the three scores is lowest, so a building is worth most when it shores up your weakest metric. Flat means the bonus stacks on top of whatever the metric already earns from food stocks, fuel reserves, or wall strength — it's an addition, not a multiplier.

One thing to watch on levels: the bonus is the value for that level, not cumulative. A Farm at level 2 grants a flat 65 to food support — not 30 + 65. Leveling up replaces the previous bonus with the higher one.

The three buildings

There are exactly three buildings, one per metric. All values below are from buildings.json. (VillageManager holds a fallback copy of this same data in BUILDING_DEFS, used only if the JSON file is missing — the JSON is the tuning surface.)

Each building has three levels. costs[0] builds level 1, costs[1] upgrades to level 2, costs[2] upgrades to level 3. bonus[i] is the flat support at that level.

Farm — food support (champion: Elder Maren)

Longhouse — shelter support (champion: Bjorn the Carpenter)

Watchtower — safety support (champion: Guard Captain Siv)

All three buildings share the same bonus ladder (30 / 65 / 110). What separates them is the material cost and which metric they feed. Want one metric cheaper or more powerful to shore up? Change that building's costs or bonus array on its own.

How often the bell offers a building

The bell rings — offering a building choice — when the village crosses a population milestone, or on a fallback timer if population growth stalls.

Three buildings, three levels each, is nine total upgrades — matched by nine population milestones plus the fallback timer, so a long-lived village can fully upgrade all three. The milestones control the pace at which that happens.

Class landmarks (champion-only)

Separate from the bell schedule, each class unlocks one signature landmark once its champion reaches skill node 3. These are built from the craft menu (not offered by the bell), gated by unlock_requires_champion_skill + unlock_requires_skill_node in recipes.json, and each is capped at 1 placement (the Bulwark allows 2). Four are village-side buildings and are covered here; two (the Miner's Deepdrill and the Smith's Grand Anvil) are machines — see Machine Tuning. Full skill/class context is in Skills, Classes, and Handicaps.

All the tuning values below live in machines.json (the landmark defs), even for the buildings.

Smokehouse — Hunter (food)

A 2×2 fuel-burning building that cures passive food, the Hunter's answer to the hunting lodge but hands-off. It banks food up to a target, then idles to save fuel.

Bulwark — Mason (defense)

A 1×1, no_fuel structure placed on the wall line (not the village), max 2. It auto-roots the wall columns in its stretch for free, and while the Maw attacks it braces its covered columns so the Maw chews through them slower. Bracing runs on a stamina duty-cycle: it braces for a stretch, then must rest at full chew-rate before recharging.

Roots persist offline; the brace slowdown is live-only (the duty-cycle needs the running Maw tick). Raise the brace mults to make a Mason wall much harder to breach; shorten rest_seconds (or lengthen brace_seconds) to raise the duty-cycle uptime.

Mead Hall — Chieftain (all metrics)

A 3×2, no_fuel building. While it stands it lifts every support metric — food, shelter, and safety — by a flat amount, so it always shores up whichever is currently lowest.

Because village health tracks the lowest metric and the bonus hits all three at once, +45 everywhere is worth more than a single bell-building level (30 / 65 / 110 to one metric). Treat it as a high-impact lever and keep it flat — support already compounds through population.

Trading Post — Merchant (gold)

A 2×2, no_fuel building that trickles passive gold, day and night, scaling with village size.

Gold accrues live and on the idle clock (offline it's idle-scaled, mirroring the live tick). Because it multiplies by population, it grows naturally with a Chieftain-less but well-fed village — raise it and gold stops being a bottleneck for the endgame exotic buys and cosmetic sinks.

Safe to change / handle with care

• The lowest metric is what matters. Overall village health tracks the weakest of food/shelter/safety, so a building bonus only helps if it lifts the metric that's currently lowest. When you retune bonus values, think about which metric a player struggles with at that stage, not the raw number.
• Keep milestones clear of the 50s. Every building_milestones value is spaced away from multiples of 50 on purpose, to avoid overlapping the population-growth cutscene. If you edit the list, keep a comfortable margin from 50/100/150/… or two cutscenes will fight for the screen.
• Gold costs gate the higher levels. Levels 2 and 3 all require gold, a traded or earned resource rather than a mined one. Raise gold costs and upgrades come later; drop them and a cash-rich village powers straight up the ladder.
• The class landmarks are champion-only power. Each lands at skill node 3 for one class, so their tuning shifts how strong a given calling feels rather than the baseline run. The Mead Hall (all metrics) and Bulwark (defense during a surge) are the two with the widest reach — move them in small steps.

After editing buildings.json, machines.json, or the village block, re-run the data validator and smoke test — see The Safe Change Workflow.

Related

• Machine Tuning
• Skills, Classes, and Handicaps
• The Villager Economy
• Villager Tasks and Projects
• Survival and Supply Drain
• The Astrolabe Consequence Curve
• Where Balance Lives