Classless
You are an Adventurer, and that is the whole job title. Identity comes from the skill web and the gear you assembled — never from a class you picked in the first five minutes and then lived with.
The forge went cold. The mountain remembers why.
A dark-fantasy classless idle RPG. Descend the mountain, rebuild the town at its foot, reignite the forge.
This is a reference, not a pitch. The game is in development and there is no public build yet — nothing here can be downloaded or played. The four labels below describe what exists in the working build, not what is available to you. Act I is the part that is finished, and nothing on this page is dressed up to look further along than it is.
You wash ashore with a crate and no plan. Above the beach there is a ruined mine town at the foot of a mountain called Azhar, and under the town there is a forge that has been cold long enough for people to stop talking about why. Everything after that is the work of finding out.
Enjin Forge II is an idle game in the sense that it keeps running while you think, not in the sense that it plays itself. Combat is zero-input — no rotations, no cooldown juggling, no ability bar. Your fights are decided before they start, by what you carried in: your gear, your skill web, the weapon you chose to master. Losing means you were underprepared, and the game tells you so in plain language rather than hiding it behind a level requirement.
The other half is the town. Ore is mined, never dropped — monsters carry monster parts, and metal comes out of rock you swung a pick at. The forge, the crusher, the mill, the kiln and the stonecutter are machines you rebuild and then feed, and they keep working while the game is closed. Rebuilding the settlement is not a side activity that pays out currency. It is how you get anything worth taking back down the mountain.
It is built by one person, in plain JavaScript, with no framework and no bundler. That shows up in the design: systems are small, legible, and slow to change, because there is nobody to hand the mess to.
Captures from a real save, not mock-ups. Click any of them to see it full size.
Where every session starts. The town at the foot of Azhar, the forge lit in the middle of it, and smoke going up.
Zero-input combat. The enemy, your health, a three-slot potion belt, and the log. Along the top: the depth you are in, the areas either side of it, and the day's calendar buff — here Deepdelver, +2% mastery XP. Nothing to press but Fight.
Act I at 100%. Ten cards — home, storage, the five machines, the trade post, the forge, and the Reignite that closes the act. This is the restoration screen: a list of projects, each one raised by pouring materials into it, not a map you lay out.
Heat to keep up with charcoal, an output bin that fills while you are away, and a bar on the go. Each recipe shows what it needs, what you actually have, and how far its mastery has come — and Iron Bar sits greyed out until Smithing 15.
The four ladders in one place — gathering, production, weapons, consumables — every recipe and every node with its own track, and rungs waiting at 25.
The shop, and the closest thing the game has to a second character — you can ask him something. Prices in silver and copper, and the day's specials sitting at the bottom with a few left.
Two gathering professions live and five still to come; six production buildings standing and two still to come. The chains are stated outright — Ore → Dust + Gems, Logs → Planks + Resin.
Three veins open in Act I and six more locked behind the acts below it. Every node carries its own mastery, so the vein you have been hitting all evening is also a ladder.
Pan and zoom. Defence, offence, utility and the tech cluster, all radiating from the free Awakening node at the centre. Respec sits in the corner, free today.
Creatures, items and lore, filled in by meeting them. Each card carries what the thing drops, how many you have killed, and which depth it belongs to.
Seven commitments the game is built around. They are not aspirations; they are the constraints every new system gets checked against, and they have survived every rework so far.
You are an Adventurer, and that is the whole job title. Identity comes from the skill web and the gear you assembled — never from a class you picked in the first five minutes and then lived with.
Combat runs itself. There are no abilities to mash — the statistics screen tracks Skills Cast and it is designed to read 0 forever. Depth lives in what you bring to a fight, not in how fast you press during one.
No billions, no trillions, no scientific notation creeping into the damage log. Numbers stay small enough to hold in your head for the entire game, because a number you cannot compare is not information.
Anything that makes you stronger is declared to your Power Rating. Economy stats — gold find, experience bonus — sit in a separate lane and are never required to progress. Nothing makes you secretly stronger or secretly worse.
Enjin is blue: ore, veins, masterwork glow. The forge and every action are ember-orange. Gold is ornament and nothing else. A colour never means two things, which is why the interface can be read at a glance.
One currency, with copper as the unit of truth underneath it. Ore is mined, never dropped. The gold supply is designed to have a single source rather than a faucet on every screen, so prices can mean something.
Epics help; epics are never required. Reliable progress rides on common, uncommon and rare drops. The rarity roll happens at drop time and is decoupled from item tier, so the thrill of a lucky drop is reachable at every level rather than only at the top.
The campaign is a descent. You start on the surface — a stretch of grey coast and the woods above it — cross to the town at the mountain's foot, and then go down, one depth at a time, each unlocked permanently by clearing the last.
There is no level gate at the door. Any depth will let you walk in. What stops you is Crushing Defense: if your gear is too far behind the area, your accuracy and damage collapse, and the game says so directly rather than refusing entry. Being turned back is information, not a locked door.
Five acts are authored as a shape for the whole game. Only the first exists in the build — the rest are design intent, stated here so the arc is legible, not because they are close.
Nine depths of Azhar. Combat, loot, gear, and the gear-check that replaces level locks. Ends by recovering three components and relighting the forge.
Crafting and smelting come into their own; the settlement grows past machines into a place with residents. Intended to open the dungeons.
Dungeons running a deliberately different combat model — exploration, random encounters, turn-based fights against several enemies, and a party recruited from a rebuilt tavern. The largest promise in the design docs and the one with the least behind it.
An endless scaling depth for build optimisation, once there is a build worth optimising.
Deliberately not completable by campaign progress alone. Intended to require something from every system in the game.
Descriptions below are the in-game text, unedited.
The active mine front. Timber supports groan in the dark. Coal dust coats everything.
Ancient tunnel network predating the village. Something vast weaves its web in the dark.
A vast glowing crystal expanse. The walls pulse with cold light. Nothing here is natural.
A collapsed labyrinth claimed by troll-kin and things that burrow. The floor trembles.
Ruins of an ancient mining civilization carved directly into the mountain. Their golems still walk.
A forgotten underground forge district. Dead furnaces, ancient machines — and something that still guards them.
Spectral depths where lost miners and forgotten smiths wander without rest or purpose.
The deepest known mines. The walls pulse with Enjin energy so intense it blinds.
The mountain's beating heart. A colossal vein of living Enjin pulses through the mountain's core. Every fragment ever mined traces its origin to this place.
Clearing all nine depths is not the ending. Three components have to be recovered from key bosses — the Ancient Bellows, the Heartstone Core and the Forgefire Ember — and carried back up to a forge that has been cold for a long time. Relighting it closes the act.
Grouped the way the game groups them in its own menu. Status labels are read out of the build, not written by hand.
Everything that happens below ground.
Nine authored depths with their own enemies, arrival text, clear quotas and resources. Clearing a depth is permanent. Bosses are an explicit choice at the quota rather than an automatic gate — the regular enemies keep spawning if you would rather keep grinding.
Swing timers, regeneration, critical hits, death and respawn, all running on their own. You choose the fight and the gear; the fight resolves itself. Damage numbers, floating combat text and a full pause are all in.
The gear check that replaces level requirements. Your power rating is compared against the area's, and a deficit costs you accuracy and damage on a fixed curve. You are never shown the numbers — just a verdict, Undergeared or Badly undergeared. It is the reason the game has no locked doors.
A small chance, after an ordinary kill in the mountain, that something other than another monster is waiting. Table-driven, so more encounters are a data change rather than a code change.
Flagged off in the build. Designed as a different mode entirely — turn-based, multi-enemy, party-based — rather than more of the same combat in a new room.
Flagged off. High-stakes encounters intended to sit behind dungeon and abyss progression.
Flagged off. The endgame idea: infinite scaling depth, no authored content, purely a measure of how good your build actually is.
What you carry, and what you have learned.
Every stat on the sheet is derived from equipment, skill web and masteries rather than stored, so nothing can silently drift out of sync with its source.
Gear is generated from a power budget across sixteen slots and nine tiers rather than hand-authored piece by piece, so the curve stays honest. Anything above an ordinary common also rolls an individual quality value that sets its item level — two swords of the same name are not the same sword.
Bags, stacking, two-handed conflict rules, and a drag-and-drop vault with sorting and deposit-all. Machine recipes count what is in storage, not just what is on you, so a full bag never stops a job.
Four ladders, not one: every gathering node, every production recipe, every weapon type and every dish carries its own mastery track to 99, with rungs at 25, 50, 75 and 99 that cut action time or add a bonus-output roll. Doing the same thing for a long time is a progression system, not just a wait.
Kills are counted per weapon type and levels are awarded, and the ladder already gates what you can forge at the Enjin Forge. The wider prestige and access layer designed on top of it does not exist yet.
A pannable, zoomable web of around fifty nodes with real effects, draft allocation and cascade removal. Respec is free once a day and costs silver after that, so changing your mind is cheap but not weightless. The engine is complete; the layout is an early one rather than the full designed web, and a few node effects are stored but not yet read.
Six machines, one settlement, and the reason to come back up.
Rebuild the settlement project by project — home, storage, the machines, the trade post, and finally the forge itself. Each is a card you open: pour gathered wood, stone and rope into it against a live requirements bar, and complete it once the bar is full. Costs are materials you mined and cut, never gold. Progress is banded by act, and only Act I is open in this build.
Smelting ore into bars, with a heat system, a hopper and fuel to manage. Unlocked once the forge foundation is restored — and relighting it is what closes Act I, so every other project in the town is on the way to this one.
Nine weapon types, each forgeable at any bar tier you have the mastery for, costed in bars and gated by your weapon mastery — including a masterwork rung at the top of each ladder. This is where mastery stops being a number and starts being access.
Crushes raw ore into dust. Reserve-on-start, refund-on-stop job queues and an output bin, like every machine in town.
Saws felled logs into planks. Its own experience curve, its own mastery, its own queue.
Cuts raw stone into dressed slabs for the building projects that need them.
Burns logs slowly into charcoal — the fuel the forge runs on.
Cooks monster scraps into simple food. As the game itself puts it: a campfire can only do so much.
The shop, and the closest thing the game has to a second character. Categories, a rotating daily special, buyback, and a stock tier that improves as you rebuild the trade post.
Designed as the single place gold is minted, by pressing rare pure gold ore into coin. The currency it would feed is live; the mint itself is a document.
Flagged off. Working code exists but no route reaches it, and the design has not been settled.
The long record.
Daily, weekly, monthly and yearly quests with their own refresh cycles, plus a multi-chain story engine. The descent itself is a quest chain, read out loud by Milo, depth by depth.
Creatures, items and lore, discovered by encountering them, with a flip reveal and filters. Kept per save file — a new game starts the collection over.
Around fifty entries awarding achievement points, claimed automatically. The milestone system was folded into this one rather than kept alongside it.
A permanent lifetime record — kills, gathering, crafting, gold, playtime — with the per-weapon and per-profession rows generated from the registries rather than hand-listed. Including, permanently, Skills Cast: 0.
Every machine keeps working while the game is shut, up to an eight-hour cap, depositing through inventory to storage to overflow so nothing is ever lost to a full bag. Combat is deliberately not included — crediting kills while away would bypass the gear check and make closing the game strictly safer than playing it.
Each weekday carries a themed economy buff, read by loot, gathering, mastery and village contribution alike. Deliberately power-neutral: it greases the economy, it does not make you stronger.
Upgrades that make a machine yield something extra — gems from crushed ore, resin from milled logs — and they work online and offline alike. Two collectors exist against a longer designed list, and both are currently unreachable: buying one costs tech points, and nothing in the game grants tech points yet.
Two kinds: gathering professions that pull material out of the world, and production professions that turn it into something else. Each has its own experience curve and its own mastery.
The table below is generated at build time from the game's own profession registry, so it cannot fall behind the build the way a hand-written list does. Anything marked as not committed is declared in the registry with no content behind it yet.
| Profession | Kind | Chain | Status |
|---|---|---|---|
| Mining | gathering | — | In the build |
| Woodcutting | gathering | — | In the build |
| Fishing | gathering | — | Not committed |
| Farming | gathering | — | Not committed |
| Smithing | production | Ore → Bars |
In the build |
| Sawmilling | production | Logs → Planks |
In the build |
| Masonry | production | Stone → Slabs |
In the build |
| Charring | production | Logs → Charcoal |
In the build |
| Crushing | production | Ore → Dust |
In the build |
| Cooking | production | Scraps → Food |
In the build |
| Alchemy | production | — | Not committed |
| Runecrafting | production | — | Not committed |
| Hunting | gathering | — | Not committed |
| Herbalism | gathering | — | Not committed |
| Excavation | gathering | — | Not committed |
Everything that has been written down, sorted by how real it is. No dates appear anywhere on this page, because there are none.
The tiers describe commitment, not timing. Something in the first tier is being worked on now; something in the third is an idea that has survived being written down, which is a much weaker claim than it sounds.
Being built right now, or built and awaiting verification.
A surface prologue before the first descent — you wash ashore with a crate, cross the coast and the woods above it, and reach the town having already used most of the things the game will ask you to use.
The merchant talks you through the opening and you answer him, ending with naming the settlement you are about to rebuild. The name you choose is then used everywhere the game refers to the town.
Short anchored explanations that appear the first time something bites you — an enemy type, retreating, dying, the potion belt — and never appear again.
Quests can now be story chains with area steps and boss-drop components. The nine-depth descent was rebuilt as one.
Stepping between depths from inside combat rather than backing out to a menu. Built, pending verification.
Levelling past roughly 28 is currently unreachable — the curve outruns any available source of experience. The replacement is verified but not yet in.
The largest single programme in flight. Remaining ability scaffolding becomes passives; auto-equip is replaced by a real side-by-side compare and a "your power against this area" readout; the talent tree goes fully passive with free respec; the gear catalogue is derived from a power budget with set bonuses; and the new-game flow is rebuilt end to end.
Design settled and written down. No implementation yet.
A one-time expensive upgrade on a mining node so it starts yielding pure ore, which smelts one-to-one instead of two-to-one. The pure ore items already exist in the game as inert glowing variants, waiting for the mechanic that gives them a reason.
Gems, leathers and resin finally get a source, which closes several recipes that currently point at items you cannot obtain.
Mastery levels currently pay very little. The fix is to make them grant access — new recipes, new nodes, new rungs — rather than invisible percentages.
The live web is an early layout. The designed one is a larger constellation carrying the stats that gear deliberately excludes, so the two systems stop overlapping.
Flat health nodes become a percentage of maximum health, so the defensive branch stops being worth a rounding error at high level.
Discovering creatures and items starts unlocking achievements and story. Explicitly power-neutral — collection never becomes a combat requirement.
A permanent in-game reference explaining each mechanic, so the one-time lessons have somewhere to live afterwards.
A full calendar panel, festivals on real dates that stack with the weekday themes, and eventually content that only opens on certain days.
Notable moments land on the screen with some weight instead of appearing as another notification in the corner.
Panels and cards get a forged finish — framed plates rather than outlined boxes — and every interactive thing gets a visible, fast response. The stated goal is that the interface stops reading like a spreadsheet.
Hardening after an unclean shutdown zeroed both a save and its backup. Unglamorous and near the top of the list.
Nothing here is promised. These are ideas that have survived being written down. Most of them will not ship, and the ones that do will not look like this.
The four unbuilt acts. The town becomes a living place with returning residents; dungeons arrive with an entirely different combat model and a recruited party; the abyss and the raids follow. The design documents describe the shape and almost none of the detail.
Fishing, farming, alchemy, runecrafting, hunting, herbalism and excavation are declared in the registry with no content behind them. Several would need entirely new places to exist — the woods and the water — before they could mean anything.
Sockets in gear, and a profession that fills them. Named in the vision document as the headline feature of that whole direction, and estimated there in months.
Hides, rune fragments and cores currently exist mostly to be sold. Giving them a real destination is an open problem, not a plan.
Each profession as a self-contained page with its station on it. Explicitly parked as an idea for some later build.
Design decisions that are genuinely unresolved. They are listed because a reference that only shows the settled parts is not much of a reference.
The rarity roll is deliberately decoupled from item tier so a lucky drop is possible anywhere. Restricting epics to bosses would make them feel earned and make the pillar about loot never walling you easier to hold — at the cost of the thing that makes every kill worth watching.
Gathering-focused and combat-focused loadouts would make the profession half a real build decision. It also risks turning every trip to the mine into an inventory chore, which is exactly the kind of friction the game keeps trying to remove.
The character sheet already shows fire, frost, nature and shadow resistances, all sitting at zero, with nothing in the game that grants or requires them. Described in the working notes as a long-standing blocker: either they get a real source and a real threat, or they come off the sheet.
The visual work is written down as a set of separate passes, but the notes concede it is really one project wearing several hats. Splitting it keeps it shippable; treating it as one is more honest about what it is.
Two editions are planned. This is a design intention rather than a shipped feature — today there is one build, and it behaves as the first of them.
No blockchain, no tokens. This is the balance baseline: the whole game is designed to be complete and fair here, and every number is tuned against it.
An optional blockchain layer behind a single switch. Anything it grants is a bonus and never a requirement — the baseline above is the game.
Vanilla JavaScript, HTML and CSS on Electron. No framework, no bundler, no module system — every file loads as a plain global script in a fixed order. It is an unusual choice in 2026 and a deliberate one: the whole codebase stays directly readable, and there is no build step between what is written and what runs.
A single feature registry is the one source of truth for what is live. Shipping a system is one flag flip, and the same flag drives the menu, the quests, and the notices that explain why something is not available. The status labels on this page are read out of that registry when the page is built.