Air is the game's stun school and its only mobility line: the school whose whole identity is control, and every ounce of that control is invisible. Stun chance is a dice roll no tooltip explains. A stunned target silently loses dodge and hands every attacker a doubled hit chance, the biggest hidden damage multiplier in the game, and Air owns it. The breakout roll that Intelligence has powered for years is documented nowhere, and the resistance that quietly builds on every landed stun has never once been shown on a screen. Even the school's dormant half is hidden: an artifact stat that rewrites three of its spells appears on no character sheet. The latest school audit could price only four of Air's ten skills, and the half it could price dies at the endgame walls the 2018 forums already mapped. This page asks for zero new stun power. It asks for the control to become visible, answerable, and connected: a charge you can see, a counterplay window you can play around, and jobs for the mobility, summon, and shield thirds the numbers cannot even measure.
Ultima Online shipped the ancestor in 1997: a lightning ladder with a cheap bolt, an expensive finisher, and a Chain Lightning that doubled its damage and split it across everyone struck — chain math as deliberate design, though lightning stayed pure damage, zero states. World of Warcraft's shaman built the build-and-spend template: routine actions charge the lightning into visible stacks, and its most honest stun is a totem with a two-second fuse anyone can kill. Guild Wars 2 made critical strikes the engine of the air rotation and drew its stun as a ring on the ground the enemy chooses to cross. Path of Exile made shock a capped, visible damage amp instead of a stun, and gave lightning a permanent job in PvE. The Elder Scrolls Online shipped the cleanest teleport governor the genre has — a recast price that climbs — and its loudest cautionary tale about fusing a stun to it. Divinity: Original Sin 2 built air as a two-step ladder where Shocked warns before Stunned lands. Baldur's Gate 3 needed one word, Wet, to organize entire parties around lightning. The verdict is consistent: nobody ships raw stuns anymore; the stun is earned through a state the victim can see and answered by a window the attacker respects.
Measured against that bar, Shroud's Air magic is mid-table and unreadable. The latest balance audit could price only 4 of its 10 combat skills, and the half it could price sits 6th of 9 magic schools at a healthy 2.78 sustained — this is not a weak school, and this page does not pretend otherwise. The gaps cluster in three places:
The deal that runs through all twenty-four proposals: zero new stun power. A stunned target silently loses all dodge and hands every attacker a doubled hit chance, so every stun second is a party-wide damage multiplier — which is exactly why stuns cannot be handed out, and why this page spends its budget on visibility, counterplay, and connection instead. Tier 1 turns the lights on: sounds, timers, honest tooltips, and the seven defects this review filed. Tier 2 builds the visible control economy — a charge you can watch build, a converter that spends it, a counterplay window both sides can play around — and gives the summon, shield, and mobility thirds real jobs. Tier 3 is the storm. One narrow exception to the stun rule exists, on A17, and it is gated, argued, and labeled where it stands.
| # | Proposal | Tier | Size | Server | One-line pitch |
|---|---|---|---|---|---|
| A1 | Thunder you can hear | 1 | S | none | every Air stun lands in silence today; the sounds the school already owns get wired |
| A2 | The tooltip confesses | 1 | S | none | stun odds, the artifact-only line, and the 2021 armor rule finally appear in the words |
| A3 | A stun you can count | 1 | S-M | none | stuns show on the target bar with a ticking timer and a combat log line |
| A4 | The ladder was always there | 1 | S-M | none | the Intelligence breakout and the resistance ladder step into the light |
| A5 | A spec for the drawn deck | 1 | S | none | Air specialization finally touches a cooldown, on either deck style |
| A6 | Discharge tells the truth | 1 | S | none | a two-second debuff beside thirty-second siblings gets aligned or documented |
| A7 | Pin the lightning down | 1 | S | none | every Air formula pinned in tests before anything gets tuned |
| A8 | The wind finally pushes | 1 | S | none | Gust's push starts pushing, with its immunities and safety rules written down |
| A9 | A ward with its own face | 1 | S | none | the ward pair stops wearing each other's icons and gets a sane unlock |
| A10 | The potion's fine print | 1 | S | none | the specialization elixir stops multiplying Lightning's range by eight |
| A11 | Wild Lightning lives | 2 | M | none | the artifact-only state that rewrites three spells becomes earnable through play |
| A12 | The charge builds | 2 | L | one stat | a visible Charged state on targets; part of the stun dice becomes a readable ramp |
| A13 | The release | 2 | M | none | a new spell spends the charge: focus back, an interrupt, a longer chain |
| A14 | Grounded | 2 | L | one stat | a visible immunity window after every stun; stronger control stops losing to weaker |
| A15 | Shield of Air: the owner's choice | 2 | M | open question | four community-argued options for the shield, and the owner picks |
| A16 | Second wind | 2 | M | none | the summon trickles focus back to its caster, the way the water summon heals |
| A17 | The elemental gets a job | 2 | M | none | the summon's authored stun proc gets repaired and the wild kin's toolkit added |
| A18 | Dash, uncaged | 2 | S-M | none | Dash breaks roots and snares, and spam taxes itself instead of a cooldown |
| A19 | Blink, sharpened | 2 | M | none | Blink stops fizzling on the move and says why a destination failed |
| A20 | Waiting for rain | 2 | M | dormant | lightning learns a conductivity term that waits, dormant, for Water's chill |
| A21 | The chain learns to count | 2 | S-M | none | the chain's damage falloff finally covers all six targets |
| A22 | Combos for the combo-dead | 2 | M | none | Air joins the combo table, Wind Slash included |
| A23 | The storm remembers | 3 | L | flagship | the storm builds charge, aims at the charged, and lends your hand a cheap strike |
| A24 | Tempest | 3 | M-L | one skill id | a capstone stance: longer chains, faster charge, and an emptying focus bar |
Picture the fight the school was built for: a bandit camp, two enemies left, an air mage opening with Lightning. The forums described this exact fight years ago — Arkah EMPstrike documented the pure-air build in 2018, and Antrax Artek spent that same year charting what casting in armor costs the school. What neither thread could describe is everything the screen refuses to say while the fight happens. A stun lands below; watch what the game shows for it.
Every Air cast runs the same road, and its last three stages are invisible to everyone in the fight:
flowchart LR A["Glyph played\nfrom the deck"] --> B["Cast gauntlet\nrange, sight, focus,\nreagents, fizzle"] B --> C["Effects fire\ndamage, push, debuff,\neach on its own roll"] C --> D["Stun roll\nchance minus the target's\nstun and control resistance"] D --> E["The stun lands\nno timer, no sound,\nno number anywhere"] E --> F["Resistance stacks\nevery landed control effect\nbuilds it, unseen"] F -.->|"feeds the next roll"| D style D stroke:#a33d2e,stroke-width:2.5px style E stroke:#a33d2e,stroke-width:2.5px style F stroke:#a33d2e,stroke-width:2.5px
Stages four through six are a complete, working control economy — a roll that reads your investment, a landing that changes the fight, a resistance that grows, a breakout that quickens — and every part of it ships in the dark. The forums have argued about Air stuns for years, in both directions, without either camp ever seeing the machine they were arguing about. The Tier 1 cards operate on exactly these stages: disclosure first, then the numbers.
Lightning is the genre's favorite element and its most argued crowd control, so the record here is unusually rich: eight eras of designers answering the same question — how do you make the strongest effect in the game fair to be hit by? The method below follows this series: measure the school against its own engine, then against the genre, then propose mostly data. Fire was rich and unreadable; Earth was solid and inert. Air's variant of the disease is stranger: the school works, sits mid-table, and hides every interesting thing it does from both sides of the fight.
Ultima Online: the words-of-power lineage this game descends from priced its lightning as a ladder — a cheap bolt, an expensive finisher, and a Chain Lightning that doubled its damage and split it across everyone struck. Chain math as a design statement, decades before anyone said "falloff table"; A21 asks this game to make the same statement on purpose. World of Warcraft's shaman is the strongest transferable pattern for a deck game: Maelstrom Weapon turned routine actions into visible stacks that a chosen cast spends — the direct model for A12 and A13 — its Capacitor Totem made a stun honest with a two-second killable fuse, and Lightning Shield made defense bite back, the model for the shield hook on A15. Guild Wars 2 made critical strikes the engine of the air rotation, the grant events A11 borrows; drew its stun as a boundary the enemy chooses to cross, the best counterplay crowd control in the genre; and its Overload burned the school's own future for present burst, the shape A24 takes. Path of Exile made shock a capped magnitude printed on the debuff icon instead of a stun — the printed-odds standard A12 holds itself to — and Arc pays more damage per remaining chain, the philosophy option A21 puts on the table; its cautionary tale is threshold math nobody can read. The Elder Scrolls Online shipped the ramping recast price A18 and A19 copy, the ally-activated storm synergy A23 adapts, and the genre's loudest warning about fusing a stun to a teleport — a warning this page obeys to the letter. Divinity: Original Sin 2 built air as a two-step ladder where Shocked warns before Stunned lands — the readable ramp A12 adapts and the visible window A14 completes — and Magicka's canon rides beside it: every player alive already believes water conducts lightning (A20). Baldur's Gate 3 needed one word, Wet, to organize whole parties around lightning, and its Call Lightning is the deck-native storm grammar A23 borrows: cast once, then harvest cheap strikes while it lives.
| Capability | UO | D:OS2 | BG3 / D&D | GW2 | WoW shaman | PoE | ESO | SotA today | SotA proposed |
|---|---|---|---|---|---|---|---|---|---|
| Visible CC state on targets (timers, marks) | ○ | ● | ● | ● | ● | ◐ | ◐ | ○ | ● A3/A12 |
| CC counterplay ladder (resist growth, immunity, telegraph) | ○ | ● | ● | ● | ● | ◐ | ● | ◐ hidden† | ● A4/A14 |
| Lightning as a damage-amp ailment | ○ | ◐ | ○ | ○ | ◐ | ● | ◐ | ◐ buried | ● A2/A6 |
| Elemental synergy states (wet, conductivity) | ○ | ● | ● | ◐ fields | ○ | ◐ | ○ | ○ | ◐ A20 waits |
| Chain behavior depth (hop rules, smart preference) | ● | ● | ◐ | ◐ | ● | ● | ○ | ◐ broken tail | ● A13/A21 |
| Mobility verbs, count and quality | ◐ | ● | ◐ | ● | ◐ | ◐ | ● | ◐ two, taxed | ● A18/A19 |
| Caster self-state engine (stacks, overload) | ○ | ○ | ◐ | ● | ● | ◐ | ◐ | ◐ artifact-only | ● A11/A24 |
| Summon with a school-specific job | ◐ | ◐ | ◐ | ◐ | ● | ● | ◐ | ○ | ● A16/A17 |
| Endgame viability against resist walls | ◐ | ● | ● wet negates | ● | ◐ | ◐ tools | ● | ○ 0.04 vs undead | ◐ balance owns |
| Defense that converts to offense | ◐ | ○ | ○ | ● | ● | ○ | ● | ○ | ● A15 |
● native · ◐ partial or era-limited · ○ absent. "SotA today" = the current development branch, 2026-08. † The half-filled cell is the page's strangest finding: a resistance ladder already answers stun spam here, building on every landed control effect and every successful breakout, and it has never been visible to either side of a fight. Surfacing it is A4; the telegraphed immunity window on top of it is A14. Row 3's "buried": the bolts already shave the target's air resistance in a stack nobody can see; A6 makes the three bolts agree on it and A2 admits it in words, while the damage-amp identity itself stays deliberately modest — Fire owns detonation. Row 9's numbers: sustained output is 0.043 against tier-5 undead and 0.697 against tier-8 elementals; the fix is the balance program's committed work, and this page links to it rather than duplicating it.
Tier 1 is the honest air: repairs and disclosure, almost all of it data and client work, published with what was broken named in player terms. Tier 2 builds the control economy the summary promised — the charge, the release, the window — and gives the summon, shield, and mobility thirds real jobs. Tier 3 is the set piece. Every tuning number below is an illustrative starting point: final values ship through the combat balance program's test-shard changesets and its measurement windows.
Every Air stun in the game lands in silence. The bolt strikes, the target freezes, and no sound plays at all — the stun impact effect simply carries no audio. There is even a finished chain lightning cast effect, with its sound authored and attached, sitting unused in the game files while a silent version plays instead. This card wires up the sounds Air already owns: stuns get an impact crack, chain lightning gets its authored audio, Blink gets an arrival sound, and the leftover orphaned effects are either connected or retired. Filed as #4407.
Seen in: WoW and FFXIV, where every crowd control effect has a signature sound and veterans track fights by ear.
Reference wiring over existing recordings — no new audio is commissioned. The pass includes an audit of the school's effect files for orphans and borrowed pieces, the same sweep the Fire page ran on its own school, with each candidate checked before it is referenced.
Air's tooltips describe a different game than the one that ships. Lightning and Chain Lightning have been unblockable, unparryable, and armor-piercing since 2021, and no text says so. Three spells describe Wild Lightning bonuses that only a rare wand family and one imbued gem can ever switch on, with no hint that the line is artifact-only (A11 is that story). Chain Lightning's own words undersell it: the live spell reaches six targets in total. Shield of Air's absorb wording does not match how the absorb actually works. Stun chances are dice rolls no tooltip explains. This card is a full truth pass: the words finally match the spells, and every stated number is checked against a pinned formula before it ships.
Seen in: Path of Exile's ailment tooltips after its clarity push; D:OS2, where every status writes its exact rules on the icon.
A strings pass, coordinated with the translation pipeline so every language receives the corrected text together, and cross-checked against the A7 pin suite so no tooltip can state a number the formulas do not.
When you stun a monster, nothing on its health bar says so, or says for how long. When a monster stuns you, the same. Stuns get a visible icon with a ticking duration on target bars and nameplates, and the combat log records the stun landing, its length, and its expiry. The strongest effect in the school stops being a secret — and the display surface this card builds is the same one the Charged pips (A12) will live on later, which is why it ships first.
Seen in: WoW, ESO, and FFXIV, where crowd control duration bars are the basic literacy of group play.
Pure presentation: the timer reads the same state the target's machine already owns, so nothing about the stun itself changes and the display can never disagree with the effect.
Two counterplay systems have shipped for years without anyone being told. Mashing to escape a stun works, Intelligence drives it, and every failed attempt shortens the remaining stun. And every landed stun or knockback quietly builds crowd control resistance on the victim, stacking up to ten times, so spamming control into the same target already gets worse and worse. Neither has ever appeared in the UI. This card surfaces both: a struggle prompt with visible progress while stunned, and the resistance stacks shown as a real buff with a real tooltip, on both sides of the fight. Nothing is invented here — the ladder exists; formalizing it into a full immunity window is A14, and this card is its display layer.
Seen in: ESO's break-free prompt; League of Legends' visible Tenacity; D:OS2, where the path from Shocked to Stunned is readable before it happens.
Air specialization never touches a single cooldown, on either deck style. The Earth page found the drawn deck missing its cooldown scaling; Air is worse, because no Air cooldown scales with mastery anywhere, locked or unlocked. Water's Cone of Cold already ships the pattern inside this same game. This card gives Air specialists the cooldown dividend every other investment of that size implies, tuned through the balance pass, with the formulas pinned (A7) before the first edit.
Seen in: this series' own Earth page, and Water's Cone of Cold in this same game.
Discharge's resistance debuff evaporates in about two seconds while the same debuff from Lightning and Chain Lightning lasts thirty. Its damage stops at five targets while its stun reaches ten. Its debuff lands unconditionally while Lightning's requires specialization. None of this reads like design; it reads like drift. This card aligns the spell with its siblings — or, where the owner prefers the current behavior, writes it down as intended. On the target caps, the recommended direction is the stun down to five, consistent with this page's zero-new-stun posture. Filed as #4404.
Seen in: the Fire page's burn-tick parity card, the same class of drift.
All data. A thirty-second debuff on a spammable spell is quietly a damage buff for the whole school, so the alignment ships through the balance program's measurement window rather than by assertion.
Before anything else changes, every Air formula gets locked in a test vise: all ten combat skills evaluated at zero mastery and at grandmaster, the chain jump table's length checked, the tree's structure audited. Every later card on this page edits numbers; this card makes those edits impossible to break silently. It is the page's first shippable artifact, and the chain-table check alone would have caught a defect that has sat in the school since 2021 (A21).
Seen in: the Fire page's expression pin suite, which unlocked that school's entire edit program.
Gust's tooltip promises to push the target back. The shipping spell is wired as a visual effect, so the push may never have actually moved anyone — possibly for years. The creature and weapon-proc versions of the same gust are wired correctly and do push. This card makes the root spell do what it says, surfaces which creatures shrug the push off, confirms it can never shove your own party, and documents its one real interaction: a knockback removes roots. Because the wiring predates the current records, the card starts with an in-game verification rather than a promise — and because a push that starts working after years changes PvP feel, it ships called out in the patch notes. Filed as #4403.
Seen in: D:OS2's Aerotheurge and WoW's Thunderstorm, where the push is the point.
The Air tree's defensive ward wears the Earth specialization's icon, and Earth's ward wears Air's, so the two most easily confused skills in both trees are also visually identical to the wrong thing. And the ward unlocks at level zero of its prerequisite, a pattern the Fire page already flagged as the odd one out. New icons for the ward pair, and a sane unlock level. The Earth page filed the shared half as #2594; this card closes the Air side of the family.
Seen in: this series' own Earth and Fire pages, which each fixed half of this family.
The Air specialization elixir secretly multiplies Lightning's range by nearly eight, out to about 156 meters, because its second bonus feeds a range stat nothing else in the game grants. Sibling elixirs feed sensible stats in that slot. This card gives the elixir a bonus that belongs there and leaves the broader elixir rebalance to the pass that owns all nine potions — the two are cross-linked, never tangled. Filed as #4405.
Seen in: no other game; this is housekeeping the school's players deserve.
Roughly half of Lightning's and Chain Lightning's authored behavior is locked behind one wand family and one imbued gem. Wild Lightning makes Lightning bounce, widens Discharge, and extends chains — and for everyone without the artifact, those branches have never fired once. Worse, the artifact swaps rather than adds: at one point of Wild Lightning the normal bolt stops firing entirely, and the wild version is blockable and loses the armor-piercing the normal bolt has carried since 2021, so the wand quietly trades away bolt quality. This card lets play grant the state, briefly, through the school's own moments — landed stuns and critical strikes build short-lived Wild Lightning — and retunes both branch families together, so the boosted path is an upgrade for everyone, wand owners included, instead of a trade nobody was told about. The wand family keeps a permanent floor the earned stacks never reach. The discoverability gap is community-documented: Asclepius asked what the gem even does, and Adam Crow and Tirrag pieced the answer together (forum thread 174373).
Seen in: WoW's Maelstrom Weapon build-and-spend, GW2's Fresh Air crit engine, PoE2's Lightning Infusion.
The state exists in the game's data today, so no new stats are needed; the grants are timed self-effects in spell data. The retune touches all three bolts in one changeset — which is why the A7 pins are a hard prerequisite, and why wand owners get a regression watch on the test shard.
Air gets a visible state on its targets: Charged. Lightning hits and landed stuns add pips you can see on the target's nameplate, with the magnitude written on the icon. Part of today's invisible stun dice moves into that ramp, so a charged target is visibly closer to a stun than a fresh one, and both sides can read it. This is not Fire's burn wearing blue: Fire's stacks are a damage ledger you detonate; Air's charge is a control ledger you convert — into focus, interruption, and longer chains (A13). Total stun output does not rise. Base chances are rebased down in the same changeset that adds the Charged term, and the rebase is measured on the test shard, not asserted. The same odds the game already rolls just become readable — and readable means answerable.
Seen in: WoW's Maelstrom stacks, PoE's shock magnitude printed on the debuff icon, D:OS2's ladder where Shocked is visible before Stunned ever lands.
One new stat on the target, appended to the table both the client and server share — append-only by design, so nothing existing can break. The state lives on the target's machine, the authoritative side for everything that happens to a character, and the card ships only after live verification with two clients: one casting, one carrying the marks, both reading the same number. The pips ride the display surface A3 builds.
A new spell spends the charge. Cast it at a Charged target and every pip converts: focus refunded per pip, the target's current cast interrupted, and your next Chain Lightning jumping one extra time per pip spent. A short, visible wind-up telegraphs the spend, so the target sees it coming. Deliberately, no stack-scaled damage — Fire already owns detonation. Air's payoff is tempo: your resource back, their spell gone, your next chain bigger, and the converter must never out-damage simply casting bolts, a constraint written into its tuning. Name candidates: Capacitance, Conduit, or Static Discharge — a name the game itself freed when the old Wand of Static Discharge was renamed to Wild Lightning in 2021.
Seen in: WoW's Maelstrom spend and old Thunderstorm mana return; D:OS2's ladder, where the payoff is earned in two steps; the Capacitor Totem's honest telegraph.
The spell rides a skill slot reserved years ago and never used, so it costs no new server work at all — the full recipe is a tree row, spell data, an icon, strings, and a test. The interrupt reuses the interruption rule the game already applies elsewhere, and the interrupt's feel in duels tunes inside the balance program's published bands.
The game already answers stun spam with a hidden resistance ladder; this card finishes the thought in the open. After a stun expires, the target gains Grounded: a brief, clearly telegraphed window of stun immunity shown right on the nameplate. And when two crowd control effects collide, the stronger one wins instead of the newer one, so a pet's weak stun can no longer overwrite a player's strong one. Air proposes this because Air owns stuns, and it is the precondition for any future conversation about stun budgets; the cross-school rollout ships through the balance pass. The history cuts both ways here: air stuns once chain-locked whole rooms in PvE, and today's PvP players complain stuns are too escapable. A visible ladder with a visible immunity window is the only design that can thread both, and the window's length is a balance-pass knob, not a number this card pretends to know. Requested by Akandriel (wishlist #778, the counterplay half; the numeric half of that ask belongs to the balance program) and answers Barugon's crowd-control stacking worry (wishlist #3982, whose gear half stays with the weapons page); the stronger-wins rule answers a standing forum ask about pets overwriting player control.
Seen in: ESO's crowd control immunity window after break-free, D:OS2's Shocked-before-Stunned ladder, and the diminishing returns every MMO ships and almost none display.
One new target-side state, appended like Charged and verified the same two-client way. The cleanest part of the design: the target's own machine applies the window the moment its stun expires, so the authority question answers itself.
The community wrote most of this card. One camp says Shield of Air's focus drain makes it a liability that caster enemies strip constantly; another says it is fine exactly as it is, provided you invest, and would rather pay focus than health. Both are on the record, so this card is a menu, not a verdict. Option one, the minimum fix: a toggle for the drain, keeping the attunement bonus. Option two: a pulsing storm aura that zaps a nearby enemy every few seconds, combat-range only so it cannot pull extra monsters. Option three: a gust shield with a chance to push melee attackers back and deflect projectiles. Option four: dispel resistance that scales with investment. And independent of the choice, an optional keystone hook: hits the shield absorbs build Charged (A12) on the attacker, so defense finally converts to offense. Worth noting: the shield's cast cost was already cut 70% in 2021 and the complaint survived, which is why this reads as a design question rather than a numbers one. Requested by Micmatty (forum thread 182343). The case for the status quo is Wishdragon's, on the record; Punkbeach argued for playstyle-flavored variants, Akandriel suggested the dispel resistance, Vaentorian the gust shield, and Duke Gréagóir weighed in. The thread's companion ask, a Shield of Focus, belongs to the Focus tree and is cross-linked there, not proposed here. Vote on this card.
Seen in: Diablo 2's lightning auras, GW2's Shocking Aura, ESO's damage shields that stay useful without draining their caster dry.
The Water elemental heals its summoner. The Air elemental, summoned by the school that spends focus faster than any other, restores nothing. This card gives it the obvious job: a periodic trickle of focus back to its caster while it fights. The summon becomes the answer to the school's real resource problem instead of a decoration, sized against the Water summon's heal. What it deliberately does not touch: pet commands and the summon rail, which belong to the taming page. Requested by Barugon (wishlist #4240; forum thread 182477).
Seen in: the game's own Water elemental, and the long MMO tradition of utility summons that feed their caster's resource.
A pet innate plus a periodic effect on the summon, with the Water heal as the sizing reference. It lands in the same balance window as whatever A15 becomes, because both touch the focus economy and the two must be measured together.
The Air elemental was authored with a stun proc that has been broken since it shipped: the effect is wired as the wrong kind, so it silently does nothing. Players noticed years ago that wild air elementals gust and shock while the summoned one never does. This card repairs the authored proc and gives the summon the wild kin's toolkit: a gust, an interrupt-flavored discharge, and scaling that already follows your own Air specialization since a recent fix. One honesty note: repairing that proc is this page's single exception to its zero-new-stun rule. The stun was authored, paid for, and merely broken — and it ships only after Grounded (A14), so the repaired stun arrives inside the visible counterplay system, never before it. Requested by FBohler (forum thread 167220, 2021); Sorgin Txakal's observation in the same thread, that summoned elementals never gust while wild ones do, shapes the kit; Wishdragon's ask for a chain-flavored area summon (forum thread 182511) is noted as a variant for the taming page's rail. Filed as #4406.
Seen in: UO's spell-casting air elementals; D:OS2's incarnates, which inherit a school's verbs; this series' own Earth page.
The mobility school's speed spell does not free you from anything: rooted is rooted, snared is snared, Dash just makes you a faster statue. Casting Dash now breaks roots and snares on use — the escape the school's identity always implied, and a rule the game half-owns already, since a knockback removes roots today. And instead of a heavier cooldown to police it, Dash borrows the cleanest governor in the genre: recasting it within a few seconds costs half again more focus each time, so spam taxes itself and honest use costs nothing extra. The ramp constant is a balance-pass knob, and the cost rule gets its own pin (A7).
Seen in: ESO's Streak, whose ramping recast cost tamed a decade of teleport spam without a cooldown; D:OS2's Uncanny Evasion.
Players asked in 2017 whether Blink was worthless: it fizzled constantly while moving, and a sheathed shield still punished the cast. The complaints were never answered. This card answers them: moving-cast fizzle relief for Blink specifically, no shield penalty when the shield is on your back, clear reticle feedback when a destination is invalid, and a small arrival payoff for specialists — a static crackle at the landing point that builds Charged (A12) on nearby enemies. Deliberately, no stun on arrival: the genre's most hated skill is the teleport with a stun stapled on. The same ramping recast governor as Dash (A18) keeps it fair, and the arrival rider ships dormant until Charged exists. Requested by Quenton (forum thread 102908, 2017); the broader fizzle complaint requested by Antrax Artek (forum thread 108426, 2018) — the full fizzle system belongs to the balance program, and this card takes only the Blink slice.
Seen in: PoE's Lightning Warp, which strikes where you land; GW2's Lightning Flash; ESO's Streak, both the governor to copy and the fused stun to refuse.
Every player alive believes water conducts lightning; fifteen years of games taught them. Shroud has no wet state anywhere, so this card is honest about what it can ship: the Air side of the wire. Lightning learns a conductivity term that amplifies damage and extends chains against cold-soaked targets, capped, and far lower in PvP than in PvE. As it happens, the Water page published beside this one proposes exactly the state to conduct through: visible chill marks that stack on a target. Frost is water; the canon holds. The term here sits dormant until Water's chill ships — this page does not invent a Water mechanic from the outside, and the two cards name each other. The school that waited years for Wild Lightning to matter knows dormant wiring can still be a promise kept.
Seen in: BG3's Wet vulnerability, D:OS2's conductive surfaces, Magicka's wet wizards; the canon rule players already know.
Chain Lightning was raised from four targets to six in 2021, but its damage falloff table was never extended, so the fifth and sixth hops quietly deal full damage where the fourth deals a quarter. The furthest strikes are the strongest, which no one designed. This card extends the table — and is honest that the fix is a nerf to hops five and six, messaged as the defect repair it is, with the returned headroom respent where the balance pass decides. Then the real question goes to the owner: what should a chain's math say? A taper that fades, the old Ultima approach of doubling the total and splitting it across everyone hit, or the modern approach where each remaining jump hits harder so short chains feel deliberate. The pin suite (A7) guards the table so it can never silently regress again. Filed as #4409.
Seen in: UO's doubled-then-split Chain Lightning, PoE's Arc paying more per remaining chain, WoW's jumps flattened after years of decay.
One of Air's fifteen skills participates in the glyph combo system, and even that one only as an ingredient in someone else's recipe. The school of connection connects to nothing. This card authors Air's cross-school entries, starting with the one a player already designed: Gust plus Double Slash makes Wind Slash, a blade carried on the wind. The Earth page's honest caveat stands here too: combos only fire from unlocked glyphs. The Earth-and-Air pairing already proposed on that page is cross-linked, never double-granted, and Water pairings wait for Water's page. Requested by Sorgin Txakal (forum thread 164727).
Seen in: Magicka's element combining, D:OS2's cross-school reactions, and this game's own Juxtapose.
Lightning Storm today is a patch of random ticks with a stun formula unlike anything else in the school — one that ignores your skill, your stats, and your specialization entirely. The rework makes the storm the school's set piece. The zone builds Charged (A12) on everyone it touches. Its bolts prefer the most charged target, with a visible strike wind-up. While your storm lives, the storm glyph in your hand becomes a cheap Strike you can call again and again — something a deck can do that a hotbar never could. Allies standing in your storm gain a brief conduit charge they can spend, so the storm mage lays fuel for the whole party. And cast the converter (A13) into your own storm, and the screen goes white. The broken stun formula is replaced in the same pass — and its repair ships early, with the Tier 1 fixes, whether or not the rework lands. Filed as #4408.
Seen in: BG3's Call Lightning, one cast then cheap strikes harvested; PoE's Orb of Storms, a zone your other casts trigger; ESO's ally-activated conduit; the Capacitor Totem's telegraphed honesty.
The zone rides the same pooled-area machinery the Earth and Fire pages already budgeted, inside the same performance caps. The ally gift is a state the zone's own tick applies, never a position the server would have to referee — which is what makes it safe in PvP. The in-hand glyph transform is the one piece that needs a feasibility spike first; if the spike fails, the storm ships without it and still works.
The capstone: a high-tier self-stance for masters only. While Tempest holds, every chain jumps one extra time and every source of Charged builds twice as fast — and the stance drains focus for every second it lives. Air's identity is potential energy, and Tempest is the decision to spend the school's future on its present: when it ends, you are empty, and everything before that was the storm. It is the one genuinely new skill this page asks for, stated as such, and it ships last — after the charge economy (A12) and Wild Lightning's stacking rules (A11) it multiplies are proven live.
Seen in: GW2's Overload Air, burning the attunement's own recharge for burst; ESO's Overload toggle spending a banked resource.
One new skill entry, the page's only one, appended to the shared skill table the same append-only way as the stats. The stance takes its own exclusive buff channel, and its drain is designed together with Shield of Air's absorb rules so the two states cannot fight each other.
The same window regions as the Section 1 screen, the same bandit fight, after the catalog ships. The bars carry the marks, the chip line carries the school's economy, the hand changes with the fight, and the forecast that was a secret is a tooltip. That is the whole page in one screen.
| Phase | Ships | Rule |
|---|---|---|
| P0 · De-risk | Nothing ships. Verify Gust's push in a live game, pin all ten spell formulas at zero mastery and grandmaster (A7), re-baseline the school audit, prove the two-client stun verification harness. | Every "already authored" claim gets proven before it is promised. |
| P1 · Repair, in the open | A1–A10, carrying the seven filed defects with them; the storm's broken stun formula gets its early repair here too. | All of Tier 1 before any new state; each fix published with what was broken, in player terms. |
| P2 · The control becomes visible | A11, A12, A13, A14. | Each state card passes the caster, target, and observer check on two live clients; A11 retunes both branch families in one changeset, never piecemeal. |
| P3 · The kit joins the loop | A15 (after the owner's pick), A16, A17 (after Grounded), A18, A19, A21, A22; A20's air-side term ships dormant. | Every card connects a third of the kit to the charge loop or answers a named community ask. |
| P4 · The storm | A23 after the hand-transform spike, A24 last; the open questions (A15's variant, A14's rollout, A20's timing) resolve from page votes and thread input. | The set piece ships after Charged is proven live; no open question resolves silently. |
flowchart LR P0["P0 verify and baseline"] --> A7["A7 pin suite"] A7 --> T1["A1 to A10\nrepairs and disclosure"] A7 --> A11["A11 Wild Lightning lives"] T1 --> A12["A12 the charge builds"] A12 --> A13["A13 the release"] A12 --> RID["A15 shield hook\nA19 arrival rider"] A12 --> A23["A23 the storm remembers"] A13 --> A23 A12 --> A24["A24 Tempest"] A11 --> A24 A4["A4 the ladder disclosed"] --> A14["A14 Grounded"] A14 --> A17["A17 the elemental's repair"] A14 --> FUT["any future stun budget"] WATER["Water page: chill ships"] -.-> A20["A20 conductivity"] style A12 stroke:#c9a55a,stroke-width:2px style A13 stroke:#c9a55a,stroke-width:2px style A23 stroke:#c9a55a,stroke-width:2px style A7 stroke:#a33d2e,stroke-width:2px
This page builds on a year of shipped repairs rather than re-asking for them: Well Grounded's air resistance was doubled (2026-07-24); Lightning Storm's stun now scales with Rapid Discharge and its duration with specialization (2026-07-24); the summoned elemental levels off its summoner's own school, and wild elementals now name their element (2026-08-03); Shock Weapon shipped (2026-07-25) and the Ignite, Frost, and Shock enchant exclusivity followed (2026-08-05); respec refunds in full, and specialization parking was fixed. One piece of history frames the whole page: in 2021 the bolts went unblockable, unparryable, and armor-ignoring, and Shield of Air's cast cost was cut 70% — and the complaints this page answers survived both changes, which is why it treats Air as a design problem rather than a numbers problem.
| Claim | Source surface |
|---|---|
| A stunned target loses dodge and attackers hit twice as often | The combat math that resolves dodge against a stunned target |
| The Intelligence breakout, and failure shortening the stun | The stun-clear routine in the combat code |
| The hidden resistance ladder, both entry points | The resistance effect every landed control applies, and the breakout success path |
| Wild Lightning's carriers and its either/or branches | The wand artifact family, the imbued gem, and the three bolt definitions |
| Chains reach six targets and the falloff table stops at four | The chain spell's data against its damage table |
| Only 4 of 10 skills priceable; the 2.78 median; the 0.043 and 0.697 walls | The 2026-07-07 balance audit |
| Every Air stun lands silently | The stun impact effect, which carries no sound |
| Gust's push wired as a visual effect | The spell's effect data, against its working creature and weapon twins |
| The elixir's 156 m bolt range | The specialization elixir's second bonus, against its siblings |
| No Air cooldown scales with specialization | The school's cooldown expressions, drawn and locked |
| One skill in fifteen joins the combo table | The combo definitions across the spell assets |
| The community record | Named threads: Micmatty 182343, Barugon 182477, FBohler 167220, Quenton 102908, Antrax Artek 108426, Sorgin Txakal 164727, the Wild Lightning thread 174373, and the 2018 build and resistance-wall threads |
The A7 pin suite is the page's first shippable artifact: every spell expression pinned at zero mastery and at grandmaster — the works-untrained invariant, numerically; the chain falloff table pinned at its full length; the breakout math pinned at low and high Intelligence; Charged accrual, cap, decay, and spend pinned as a state machine; Grounded's three states (applies when a stun ends, blocks what it should, expires on time); the converter's refund math per pip; and Wild Lightning's either/or swap pinned in both directions, so the A11 retune cannot regress silently. The one thing no automated test can cover is multi-client state sync; every stat card on this page carries the caster, target, and observer checklist on two live clients.
Shipping game data (the skill sheet, the thirteen Air spell definitions, the resistance and channel data, the wand artifact family); the combat engine's effect, formula, and targeting code; the balance audit and change log; the 2025–2026 wishlist sweep and the named forum threads; and the genre record — Ultima Online's lightning ladder and chain math, D:OS2's Aerotheurge, BG3's Wet and Call Lightning, GW2's Fresh Air and Static Field, WoW's Maelstrom and Capacitor, PoE's shock and Arc, ESO's Storm Calling, with Magicka's conductivity canon.
No account needed. Every proposal above also has its own Discussion box for talking about that one alone; this thread is for the plan as a whole. Comments are plain text, held to basic decency, and may be trimmed.