OpenRA-Bench — Scenario Catalog

Two regions, 6 units. Put >=2 units in EACH region (NE ~110,6 and SE ~110,33) before tick 2400. One direct parallel dispatch makes it (~17-20 turns); touring both with one column does not. Lose if the deadline passes or the force is wiped.
WIN WHEN: get ≥2 units into region (110,6) r=8 AND get ≥2 units into region (110,33) r=8 AND before game tick 2400.
YOU LOSE IF: (your whole force is destroyed OR not before game tick 2401).
You have at most 30 decision turns; acting decisively and early matters.

Three DISPERSED regions, 8 units, attrition cap. Put >=2 units in EACH of: NE corner (115,6), bottom-left (20,36), SE corner (115,34) before tick 2800, losing <=2. The regions are in three different directions — a single eastward column cannot reach the bottom-left one — so this needs genuine multi-vector dispatch, not one fanned-out march. Enemy infantry contest the two eastern lanes; the bottom-left route is clear.
WIN WHEN: get ≥2 units into region (115,6) r=7 AND get ≥2 units into region (20,36) r=7 AND get ≥2 units into region (115,34) r=7 AND before game tick 2800 AND lose ≤2 of your own units.
YOU LOSE IF: (your whole force is destroyed OR not before game tick 2801).
You have at most 36 decision turns; acting decisively and early matters.

Three dispersed regions, 8 units, attrition cap — you are NOT given coordinates. Each target region is marked by a specific enemy building, hidden in fog until you scout it. Put >=2 units next to EACH of: the enemy Construction Yard in the NORTH-EAST, the enemy Power Plant in the SOUTH-WEST (bottom-left), and the enemy Ore Refinery in the SOUTH-EAST, before tick 2800, losing <=2. Push a group toward each direction and KEEP ADVANCING — do not stop at a guessed spot. The instant a unit spots a target building you will be interrupted and shown its exact location; then converge >=2 units on it. The three lie in three different directions; a single eastward column cannot reach the south-west one, so this needs genuine multi-vector dispatch. Enemy infantry contest the two eastern lanes; the south-west route is clear.
WIN WHEN: get ≥2 units into the enemy Construction Yard (fact) in the NORTH-EAST AND get ≥2 units into the enemy Power Plant (powr) in the SOUTH-WEST (bottom-left) AND get ≥2 units into the enemy Ore Refinery (proc) in the SOUTH-EAST AND before game tick 2800 AND lose ≤2 of your own units.
YOU LOSE IF: (your whole force is destroyed OR not before game tick 2801).
You have at most 36 decision turns; acting decisively and early matters.

Execute ONE given route IN ORDER: waypoint W1 (mid-south, marked by an enemy Power Plant) → W2 (north, enemy Ore Refinery) → W3 / final (far SE, enemy Construction Yard). You must reach them in that order — rushing straight to the final does NOT count. Keep the column moving: arrive within the time budget or you lose.
WIN WHEN: reach these waypoints IN ORDER (no skipping, no idling): region (40,30) r=3 → then → region (95,9) r=3 → then → region (118,33) r=3 AND before game tick 2400.
YOU LOSE IF: (not before game tick 2401 OR your whole force is destroyed).
You have at most 30 decision turns; acting decisively and early matters.

TWO given routes that must BOTH be completed, each IN ORDER. Route N: N1 (W, enemy Power Plant) → N2 (mid-SOUTH, enemy Ore Refinery) → N3 (far-E NORTH, enemy Construction Yard). Route S: S1 (W-south, enemy Radar Dome) → S2 (mid-NORTH, enemy Power Plant) → S3 (far-E SOUTH, enemy Ore Refinery). The two routes cross in the middle: split into two columns and run them in parallel — doing one fully then the other overruns the clock. Lose ≤1 unit; arrive within budget or lose.
WIN WHEN: reach these waypoints IN ORDER (no skipping, no idling): region (45,8) r=3 → then → region (90,32) r=3 → then → region (120,8) r=3 AND reach these waypoints IN ORDER (no skipping, no idling): region (45,32) r=3 → then → region (90,8) r=3 → then → region (120,32) r=3 AND before game tick 3000 AND lose ≤1 of your own units.
YOU LOSE IF: (not before game tick 3001 OR NOT (lose ≤1 of your own units) OR your whole force is destroyed).
You have at most 36 decision turns; acting decisively and early matters.

TWO given routes across a large arena, each four ordered waypoints, described ONLY by direction — you are NOT given coordinates. Each waypoint is an enemy building hidden in fog: scout toward the stated direction and you will be interrupted the instant a unit sights the marker, revealing it; then move on to it and continue the route IN ORDER. Both routes must be completed (split into two columns and run them together). Keep advancing — idling, skipping, or doing one route then the other overruns the clock. Lose ≤1 unit.
WIN WHEN: reach these waypoints IN ORDER (no skipping, no idling): the far NORTH-WEST — search the western ~quarter of the map, in the northern ~third → then → the SOUTH-CENTRE — mid map left-to-right, in the southern ~quarter → then → the EAST side — the eastern ~quarter, in the northern ~third → then → the far SOUTH-EAST corner — extreme east, extreme south AND reach these waypoints IN ORDER (no skipping, no idling): the far SOUTH-WEST — the western ~quarter, in the southern ~third → then → the NORTH-CENTRE — mid map left-to-right, in the northern ~quarter → then → the EAST side — the eastern ~quarter, in the southern ~third → then → the far NORTH-EAST corner — extreme east, extreme north AND before game tick 6000 AND lose ≤1 of your own units.
YOU LOSE IF: (not before game tick 6001 OR NOT (lose ≤1 of your own units) OR your whole force is destroyed).
You have at most 70 decision turns; acting decisively and early matters.

Two docks, one shared shipping deadline. Dock A (Barracks, type 'barr') is in the NE (~115,6); dock B (Refinery, type 'proc') is in the SW (~20,36). Put >=3 of your units at EACH dock before tick 2700: the joint AND-clause only fires when BOTH docks are staffed AT THE SAME INSTANT. Split your force and dispatch BOTH squads at once — a single column touring A then B leaves A empty when it reaches B and never satisfies the joint win. Lose if the deadline passes or the force is wiped.
WIN WHEN: get ≥3 units into region (115,6) r=8 AND get ≥3 units into region (20,36) r=8 AND before game tick 2700.
YOU LOSE IF: (your whole force is destroyed OR not before game tick 2701).
You have at most 32 decision turns; acting decisively and early matters.

Two docks, ONE tight shared shipping deadline. Dock A (Barracks 'barr') is far in the NE (~115,6); dock B (Refinery 'proc') is in the SW (~20,36). Put >=3 of your units at EACH dock before tick 2200: the joint AND-clause only fires when BOTH docks are staffed AT THE SAME INSTANT. The NE dock is much farther than the SW dock, so the NE squad must be launched FIRST (the stagger) and the SW squad must follow promptly. A single column touring A then B leaves A empty when it reaches B and cannot finish in time. Lose if the deadline passes or the force is wiped.
WIN WHEN: get ≥3 units into region (115,6) r=8 AND get ≥3 units into region (20,36) r=8 AND before game tick 2200.
YOU LOSE IF: (your whole force is destroyed OR not before game tick 2201).
You have at most 26 decision turns; acting decisively and early matters.

Three docks, one tight shared shipping deadline. You are NOT given coordinates. Each dock is marked by a specific enemy building: a Barracks (barr) in the NORTH-EAST, an Ore Refinery (proc) in the SOUTH-WEST (bottom-left), and a Power Plant (powr) in the SOUTH-EAST. Put >=3 of your units at EACH dock before tick 2700: the joint AND-clause only fires when ALL THREE docks are staffed AT THE SAME INSTANT. The three docks lie in three DIFFERENT directions — a single column cannot cover all three — so you need true three-way parallel dispatch. Lose if the deadline passes or the force is wiped.
WIN WHEN: get ≥3 units into the enemy Barracks (barr) in the NORTH-EAST AND get ≥3 units into the enemy Ore Refinery (proc) in the SOUTH-WEST (bottom-left) AND get ≥3 units into the enemy Power Plant (powr) in the SOUTH-EAST AND before game tick 2700.
YOU LOSE IF: (your whole force is destroyed OR not before game tick 2701).
You have at most 32 decision turns; acting decisively and early matters.

A single enemy rifleman is patrolling INSIDE your perimeter, bouncing between (14,20) and (30,20). Every western swing it chips your refinery (`proc` at 16,20). Your 5 riflemen are on HOLD-FIRE — they will not engage unless you order it. Keep them ON the refinery and order them to attack the probe as it swings in; let your own walls bring it to you. Win when the probe is dead, the refinery still stands, and you have not lost more than 2 of your own — before tick 4500. Sitting idle (the probe destroys the refinery) and chasing the probe off east (refinery undefended) both LOSE.
WIN WHEN: destroy ≥1 enemy units AND keep ≥4 units alive AND lose ≤2 of your own units AND own a 'proc' AND before game tick 4500.
YOU LOSE IF: (not before game tick 4501 OR NOT (lose ≤2 of your own units) OR NOT (own a 'proc')).
You have at most 55 decision turns; acting decisively and early matters.

Two enemy riflemen patrol the same arc inside your perimeter, bouncing between (14,20) and (30,20). Hold the line ON your refinery (`proc` at 16,20), order BOTH probes engaged as they swing west; do not chase them east. Win when ≥2 enemies are down, the refinery still stands, ≥5 of your team are alive and you have not lost more than 1 — before tick 4500. Sitting idle (refinery dies under double chip), chasing the probes east (refinery undefended OR team caught out), or pulling back (refinery dies) all LOSE.
WIN WHEN: destroy ≥2 enemy units AND keep ≥5 units alive AND lose ≤1 of your own units AND own a 'proc' AND before game tick 4500.
YOU LOSE IF: (not before game tick 4501 OR NOT (lose ≤1 of your own units) OR NOT (own a 'proc')).
You have at most 55 decision turns; acting decisively and early matters.

THREE enemy riflemen probe your perimeter on the central latitude (anchor 22,20 → bouncing 14..30,20). Your refinery (`proc` at 16,20) AND your adjacent barracks (`tent` at 18,20) must BOTH still stand. Your 5 riflemen stage on a seed-varied latitude (north y=18 or south y=22) — both slots cover BOTH assets. They are on HOLD-FIRE; order them to engage each probe as it swings west into your assets' footprint. Win when ≥3 probes are dead, both buildings still stand, ≥5 of you are alive and you have NOT lost a single defender — before tick 4500. Idling, chasing east, or pulling back all LOSE.
WIN WHEN: destroy ≥3 enemy units AND keep ≥5 units alive AND lose ≤0 of your own units AND own a 'proc' AND own a 'tent' AND before game tick 4500.
YOU LOSE IF: (not before game tick 4501 OR NOT (lose ≤0 of your own units) OR NOT (own a 'proc') OR NOT (own a 'tent')).
You have at most 55 decision turns; acting decisively and early matters.

Rush Hour Sweep - easy. Sweep the arena: spread your team across the scattered squads, reveal the map and destroy >=7 enemy units within the time budget, losing <=5 of your own. Sitting still, marching one straight line down a single row, or piling onto a single squad cannot reach the kill AND map-coverage bars in time - that is a loss, not a draw, and so is timing out or losing the team.
WIN WHEN: destroy ≥7 enemy units AND reveal ≥18% of the map AND before game tick 3603 AND lose ≤5 of your own units.
YOU LOSE IF: (your whole force is destroyed OR NOT (lose ≤5 of your own units) OR not before game tick 3603).
You have at most 40 decision turns; acting decisively and early matters.

Rush Hour Sweep - medium. Wider coverage and a tighter attrition cap: destroy >=9 enemy units and reveal >=28% of the arena before tick 3603, losing <=4. A single-row sweep tops out at 8 kills - clearing both the north and south bands is required. Partial or passive sweep, timeout, force loss, or busting the attrition cap is a loss.
WIN WHEN: destroy ≥9 enemy units AND reveal ≥28% of the map AND before game tick 3603 AND lose ≤4 of your own units.
YOU LOSE IF: (your whole force is destroyed OR NOT (lose ≤4 of your own units) OR not before game tick 3603).
You have at most 40 decision turns; acting decisively and early matters.

Rush Hour Sweep - hard. The team round-robins between four base spawn corners by seed (start corner varies each run, anti-memorisation), and the objective is given by compass description, not coordinates: you must localise the scattered squads on the minimap yourself. Destroy >=10 enemy units and reveal >=32% of the arena before tick 3603, losing <=4. A single-row sweep cannot reach the kill bar; only a real spread across north, mid and south clears it. Partial sweep, timeout, force loss, or attrition bust is a loss.
WIN WHEN: destroy ≥10 enemy units AND reveal ≥32% of the map AND before game tick 3603 AND lose ≤4 of your own units.
YOU LOSE IF: (your whole force is destroyed OR NOT (lose ≤4 of your own units) OR not before game tick 3603).
You have at most 40 decision turns; acting decisively and early matters.

Two-Body Coordination — you command TWO squads that start on OPPOSITE edges of the map. There are TWO enemy bases: one on the WEST (near 24,20), one on the EAST (near 78,20), each with a construction yard (fact) and refinery (proc). You must raze fact+proc at BOTH bases. The map's center is a lethal turret wall: a squad cannot cross it to reach the far base, so a single squad or a do-one-then-the-other plan cannot finish in time. Drive BOTH squads forward at once — left squad onto the west base, right squad onto the east base — simultaneously.
WIN WHEN: destroy the enemy fact+proc AND destroy the enemy fact+proc at the base near (24,21) AND destroy the enemy fact+proc at the base near (78,21) AND before game tick 8500.
YOU LOSE IF: (NOT (before game tick 8500) OR your whole force is destroyed).
You have at most 100 decision turns; acting decisively and early matters.

Same two-base objective, tighter clock and an attrition cap — only clean simultaneous control of both squads finishes in time without bleeding the force.
WIN WHEN: destroy the enemy fact+proc AND destroy the enemy fact+proc at the base near (24,21) AND destroy the enemy fact+proc at the base near (78,21) AND before game tick 6500 AND lose ≤10 of your own units.
YOU LOSE IF: (NOT (before game tick 6500) OR NOT (lose ≤10 of your own units) OR your whole force is destroyed).
You have at most 90 decision turns; acting decisively and early matters.

Both bases razed (fact+proc each), tight clock, strict attrition: both squads must strike their own base in parallel and clean. Any serialized or single-squad attempt blows the clock or the attrition cap.
WIN WHEN: destroy the enemy fact+proc AND destroy the enemy fact+proc at the base near (24,21) AND destroy the enemy fact+proc at the base near (78,21) AND before game tick 5000 AND lose ≤6 of your own units.
YOU LOSE IF: (NOT (before game tick 5000) OR NOT (lose ≤6 of your own units) OR your whole force is destroyed).
You have at most 80 decision turns; acting decisively and early matters.

Spec: build EXACTLY 3 e1 rifles + EXACTLY 1 e2 grenadier — requires erecting a barracks first (the tech prerequisite for both). The budget covers the barracks + the precise mix; a 4th e1, a 2nd e2, or any extra structure breaks the spec. Deliver before the deadline — stalling, under-delivering, or producing the wrong mix (e.g. only e1) is a loss.
WIN WHEN: have EXACTLY 3 'e1' (no more, no fewer) AND have EXACTLY 1 'e2' (no more, no fewer) AND before game tick 2700.
YOU LOSE IF: (not before game tick 2701 OR have ≥4 'e1' OR have ≥2 'e2').
You have at most 40 decision turns; acting decisively and early matters.

Spec: build EXACTLY 3 e1 + EXACTLY 2 e3 (rocket). e3 shares the barracks tech but costs more — the budget forces correct allocation, no waste. Over- or under-producing either type, or missing the deadline, is a loss.
WIN WHEN: have EXACTLY 3 'e1' (no more, no fewer) AND have EXACTLY 2 'e3' (no more, no fewer) AND before game tick 3000.
YOU LOSE IF: (not before game tick 3001 OR have ≥4 'e1' OR have ≥3 'e3').
You have at most 45 decision turns; acting decisively and early matters.

Spec: build EXACTLY 3 e1 + EXACTLY 2 e3, plus a working Tesla Coil (tsla) — a deep soviet tech path (proc → weap → tsla) that also draws heavy power, so extra power plants are required to keep a non-negative power surplus. The tight budget leaves no room for overproduction or an unnecessary structure; under-delivering or missing the deadline is a loss.
WIN WHEN: have EXACTLY 3 'e1' (no more, no fewer) AND have EXACTLY 2 'e3' (no more, no fewer) AND own ≥1 'tsla' building(s) AND keep power surplus ≥0 AND before game tick 6300.
YOU LOSE IF: (not before game tick 6301 OR have ≥4 'e1' OR have ≥3 'e3' OR own ≥2 'tsla' building(s)).
You have at most 75 decision turns; acting decisively and early matters.

Execute the GIVEN ordered procedure with only move_units / stop. Visit the three staging points STRICTLY IN ORDER: S1 (≈ x40,y30) → S2 (≈ x95,y10) → S3 (≈ x118,y33). Reaching a later point before its predecessor does NOT count and does NOT skip ahead — the order is latched. Complete the full ordered route before the deadline or you lose.
WIN WHEN: reach these waypoints IN ORDER (no skipping, no idling): region (40,30) r=3 → then → region (95,10) r=3 → then → region (118,33) r=3 AND before game tick 2600.
YOU LOSE IF: (not before game tick 2601 OR your whole force is destroyed).
You have at most 34 decision turns; acting decisively and early matters.

Execute the GIVEN ordered procedure (move_units / stop only). Visit FOUR staging points STRICTLY IN ORDER: S1 (≈ x40,y30) → S2 (≈ x95,y10) → S3 (≈ x40,y33) → S4 (≈ x118,y20). A later point reached early does NOT count and does NOT advance the sequence; a skipped point can never be recovered. Run the full route in order before the deadline.
WIN WHEN: reach these waypoints IN ORDER (no skipping, no idling): region (40,30) r=3 → then → region (95,10) r=3 → then → region (40,33) r=3 → then → region (118,20) r=3 AND before game tick 3400.
YOU LOSE IF: (not before game tick 3401 OR your whole force is destroyed).
You have at most 44 decision turns; acting decisively and early matters.

Execute the GIVEN ordered procedure (move_units / stop only) from a seed-chosen staging corner. Visit FOUR staging points STRICTLY IN ORDER: S1 (≈ x40,y30) → S2 (≈ x95,y10) → S3 (≈ x40,y33) → S4 (≈ x118,y20). Order is latched: a later point reached early does not count or skip ahead, a skipped point is unrecoverable. Complete the full ordered route within the tight deadline.
WIN WHEN: reach these waypoints IN ORDER (no skipping, no idling): region (40,30) r=3 → then → region (95,10) r=3 → then → region (40,33) r=3 → then → region (118,20) r=3 AND before game tick 3700.
YOU LOSE IF: (not before game tick 3701 OR your whole force is destroyed).
You have at most 46 decision turns; acting decisively and early matters.

Even duel: your 4 vehicles vs 3 enemy rifle infantry a short move east. Win by destroying the enemy force without losing all of yours.
WIN WHEN: destroy ≥3 enemy units AND before game tick 2400.
YOU LOSE IF: (not before game tick 2401 OR your whole force is destroyed).
You have at most 30 decision turns; acting decisively and early matters.

Tougher opponent: the enemy rifle squad is backed by light armour. Same close engagement; concentrate fire to win the trade.
WIN WHEN: destroy ≥5 enemy units AND before game tick 3000.
YOU LOSE IF: (not before game tick 3001 OR your whole force is destroyed).
You have at most 40 decision turns; acting decisively and early matters.

Move a unit to the goal zone near (55,16) at the far east of the playable area. No enemy on this map — success is purely reading the terrain and committing a route immediately. Win: a unit reaches the zone before the deadline. You LOSE if no unit is in the zone by the deadline (idling/stalling/wandering loses).
WIN WHEN: get a unit into region (55,16) r=5 AND before game tick 900.
YOU LOSE IF: (not before game tick 901 OR your whole force is destroyed).
You have at most 14 decision turns; acting decisively and early matters.

Same task, harder: the goal zone is farther at (72,16) and EVERY unit must reach it (not just the fast scout). Plan the longer route and march the whole squad directly. Win: every unit is in the zone before the deadline. You LOSE if the whole squad is not in the zone by the deadline.
WIN WHEN: get EVERY unit into region (72,16) r=5 AND before game tick 1500.
YOU LOSE IF: (not before game tick 1501 OR your whole force is destroyed).
You have at most 20 decision turns; acting decisively and early matters.

The whole squad must reach the goal zone at the far east-central edge of the playable area, but you are NOT given its coordinates — localize it on the minimap yourself. You start from a seed-chosen latitude (two spawn_point groups) so the route can't be replayed from memory. Commit a direct full-force march. Win: every unit is in the zone before the deadline. You LOSE if the whole squad is not in the zone by the deadline.
WIN WHEN: get EVERY unit into the goal zone at the FAR EAST-CENTRAL edge of the playable area (right side of the map, vertically centred — march your whole squad straight east from the spawn) AND before game tick 1600.
YOU LOSE IF: (not before game tick 1601 OR your whole force is destroyed).
You have at most 22 decision turns; acting decisively and early matters.

Move a unit to the egress zone near (55,16) at the far east of the playable area. No walls, no enemy — success is purely reading the terrain and committing the route immediately. Win: a unit reaches the zone before the deadline. You LOSE if no unit is in the zone by the deadline (idling/stalling/wandering loses).
WIN WHEN: get a unit into region (55,16) r=5 AND before game tick 900.
YOU LOSE IF: (not before game tick 901 OR your whole force is destroyed).
You have at most 14 decision turns; acting decisively and early matters.

Same task, harder: the egress zone is farther at (72,16) and EVERY unit must reach it (not just the fast scout). Plan the longer route and march the whole squad directly. Win: every unit is in the zone before the deadline. You LOSE if the whole squad is not in the zone by the deadline.
WIN WHEN: get EVERY unit into region (72,16) r=5 AND before game tick 1500.
YOU LOSE IF: (not before game tick 1501 OR your whole force is destroyed).
You have at most 20 decision turns; acting decisively and early matters.

Confined aisle map. Three vertical silo walls (x≈18, 30, 42) split the playable area into four aisles; each wall has ONE 6-cell gap that alternates top/bottom/top (wall A gap at y=4..9, wall B at y=30..35, wall C at y=4..9), so the only path is a serpentine detour. The whole squad of 4 infantry must reach the egress zone at the far south-east corner (centred near (55,30), radius 6) before the deadline, with zero losses. You start from a seed-chosen staging row (two spawn_point groups: NORTH y=8..14 or SOUTH y=25..31) so the leg lengths differ between seeds. Win: every unit in the zone before the deadline, none lost. You LOSE on the deadline, OR if any unit is lost (recklessly attacking silos to bash through wastes the budget — commit a clean detour).
WIN WHEN: get EVERY unit into region (55,30) r=6 AND before game tick 2700 AND lose ≤0 of your own units.
YOU LOSE IF: (not before game tick 2701 OR NOT (lose ≤0 of your own units)).
You have at most 31 decision turns; acting decisively and early matters.

One hidden marker building sits far east, off the scouts' spawn rows, inside a single contiguous fog mass. Reading "the unknown is the eastern dark band, drive a scout into it" is direct. Generous clock — but stalling or a blind near-spawn path never enters the pocket and loses on the deadline.
WIN WHEN: spot ≥1 enemy buildings AND before game tick 2200.
YOU LOSE IF: (not before game tick 2201 OR your whole force is destroyed).
You have at most 26 decision turns; acting decisively and early matters.

The frontier is split in two: a near pocket and a FAR one. Two hidden marker buildings must BOTH be discovered. A decoy enemy squad sits just off the spawn lane inside the already-seen band to bait scouts into re-treading known ground; the second marker is far NE and only a deliberate split push reveals it in time.
WIN WHEN: spot ≥2 enemy buildings AND before game tick 2800.
YOU LOSE IF: (not before game tick 2801 OR your whole force is destroyed).
You have at most 32 decision turns; acting decisively and early matters.

The frontier is fragmented across three corners — far NE, far SE, and a NW strip — each hiding one marker building; ALL three must be discovered. A decoy squad sits in the explored center to pull scouts off the true frontier, the spawn corner is seed-varied (so a memorised reading order cannot generalise), and the clock is short: only a correct, simultaneous read of all three fog pockets discovers every marker in time.
WIN WHEN: spot ≥3 enemy buildings AND before game tick 3500.
YOU LOSE IF: (not before game tick 3501 OR your whole force is destroyed).
You have at most 40 decision turns; acting decisively and early matters.

One target building, one unexplored region — far east, roughly on the spawn latitude. The near area is already open (no decoy fog). The bare skill: read "the fog is east, therefore the target is east" and commit a sustained eastward push. Idling, a one-cell nudge, or dithering on the nearest frontier runs out the clock and LOSES — only a committed read of where the fog mass is wins.
WIN WHEN: spot ≥1 enemy buildings AND before game tick 2000.
YOU LOSE IF: (NOT (before game tick 2000) OR your whole force is destroyed).
You have at most 26 decision turns; acting decisively and early matters.

Two unexplored regions: a NEAR one (NE, near the spawn latitude) that holds only enemy UNITS — no building — and a FAR one (SE, opposite corner) that holds the real target building. A blind eastward charge sweeps the near pocket, finds no building, and runs out the clock; the nearest-frontier sweep never reaches the far corner. Only inferring "the near fog is empty, so the target must be in the remaining unexplored region" and committing SE wins in time. Missing the deadline is a LOSS.
WIN WHEN: spot ≥1 enemy buildings AND before game tick 2300.
YOU LOSE IF: (NOT (before game tick 2300) OR your whole force is destroyed).
You have at most 30 decision turns; acting decisively and early matters.

Two decoy pockets — a NE squad and a mid-SOUTH squad, both enemy units only, no building — flank the lane to the real target, a lone building deep east on the mid latitude (counter-intuitive relative to a spawn that pushes along its own edge). Short clock and an attrition cap (≤1 unit lost): plowing scouts straight through a decoy busts the cap or the clock; only reading which fog pocket can still hold the target AND routing the clear mid-lane around the decoys discovers it in time. Two seed-varied spawn points (NW / SW).
WIN WHEN: spot ≥1 enemy buildings AND before game tick 2700 AND lose ≤1 of your own units.
YOU LOSE IF: (NOT (before game tick 2700) OR NOT (lose ≤1 of your own units) OR your whole force is destroyed).
You have at most 34 decision turns; acting decisively and early matters.

A hunting picket (rocket infantry + riflemen) blocks the lane to the objective and chases your nearest unit. Get THREE tanks to the objective (far east). Driving the column straight in gets it shot apart before three arrive; send the fast jeeps in as bait on a separate vector to pull the hunters off, then run the tanks through. Idling or never committing loses on the clock.
WIN WHEN: get ≥3 units into region (112,20) r=6 AND before game tick 5800.
YOU LOSE IF: (not before game tick 5801 OR your whole force is destroyed).
You have at most 66 decision turns; acting decisively and early matters.

Heavier rocket-infantry picket — a head-on push is now decisively fatal. Same task: divert the hunters with the bait, land THREE tanks on the objective before the (tighter) deadline.
WIN WHEN: get ≥3 units into region (112,20) r=6 AND before game tick 6600.
YOU LOSE IF: (not before game tick 6601 OR your whole force is destroyed).
You have at most 76 decision turns; acting decisively and early matters.

Strongest hunting picket; the force stages from a seed-chosen latitude (the decoy line that works varies by seed, so it can't be memorised). Land THREE tanks on the objective AND lose ≤2 units (only the two bait jeeps may be spent — burning armour fails) before the tight deadline.
WIN WHEN: get ≥3 units into region (112,20) r=6 AND before game tick 7400 AND lose ≤2 of your own units.
YOU LOSE IF: (not before game tick 7401 OR NOT (lose ≤2 of your own units) OR your whole force is destroyed).
You have at most 84 decision turns; acting decisively and early matters.

A leashed rocket-infantry wall blocks the direct lane east to the objective and shreds anything that tries to cross it. Get THREE tanks to the objective (far east, marked by the enemy structure). Driving the column straight in loses it in a turn; go the long way — climb to the open flank, run east PAST the end of the wall, then turn down to the objective. Idling never commits and loses on the clock.
WIN WHEN: get ≥3 units into region (112,20) r=6 AND before game tick 4000 AND lose ≤1 of your own units.
YOU LOSE IF: (not before game tick 4001 OR NOT (lose ≤1 of your own units) OR your whole force is destroyed).
You have at most 60 decision turns; acting decisively and early matters.

The guard wall now extends much farther east — the long way round is longer and costlier, and a single clipped tank now fails. Same task: ignore the lethal short lane entirely, take the flank route around the (longer) wall's end, land THREE tanks on the objective before the tighter deadline.
WIN WHEN: get ≥3 units into region (112,20) r=6 AND before game tick 4600 AND lose ≤0 of your own units.
YOU LOSE IF: (not before game tick 4601 OR NOT (lose ≤0 of your own units) OR your whole force is destroyed).
You have at most 66 decision turns; acting decisively and early matters.

Strongest, longest wall; the force stages from a seed-chosen latitude (north or south of the wall), so the length and shape of the long way round varies by seed and can't be memorised. The short lane is lethal from either start. Get the WHOLE force (every surviving tank, ≥3) onto the objective with ZERO losses, via the long way around the wall's end, before the tight deadline.
WIN WHEN: get EVERY unit into region (112,20) r=6 AND keep ≥3 units alive AND before game tick 5000 AND lose ≤0 of your own units.
YOU LOSE IF: (not before game tick 5001 OR NOT (lose ≤0 of your own units) OR your whole force is destroyed).
You have at most 70 decision turns; acting decisively and early matters.

A strong rocket-infantry guard wall blocks the one lane to the objective (far east, marked by the enemy structure) and snaps straight back to it the instant nothing threatens it. Get THREE tanks to the objective. Driving the column straight in runs it into the wall and it is destroyed; send the fast jeeps in close on a separate vector so a segment of the wall lunges off post after them, then run the tanks through the open slot before it snaps back. Only the two bait jeeps may be spent — burning a tank (≥3 lost) fails. Idling never commits and loses on the clock.
WIN WHEN: get ≥3 units into region (112,20) r=6 AND before game tick 5800 AND lose ≤2 of your own units.
YOU LOSE IF: (not before game tick 5801 OR NOT (lose ≤2 of your own units) OR your whole force is destroyed).
You have at most 66 decision turns; acting decisively and early matters.

Denser rocket-infantry wall — a head-on push is now decisively fatal and the bait must pull a wider segment off the lane. Same task: lure a segment off post with the jeeps on a divergent vector, run THREE tanks through the open slot to the objective before the (tighter) deadline, losing only the two bait jeeps.
WIN WHEN: get ≥3 units into region (112,20) r=6 AND before game tick 6600 AND lose ≤2 of your own units.
YOU LOSE IF: (not before game tick 6601 OR NOT (lose ≤2 of your own units) OR your whole force is destroyed).
You have at most 76 decision turns; acting decisively and early matters.

Densest guard wall; the force stages from a seed-chosen latitude (north or south), so the bait vector and the slot the main body runs through vary by seed and can't be memorised. Lure a segment off post with the jeeps, then run THREE tanks through the open slot to the objective AND lose ≤2 units (only the two bait jeeps may be spent — burning armour fails) before the tight deadline.
WIN WHEN: get ≥3 units into region (112,20) r=6 AND before game tick 7400 AND lose ≤2 of your own units.
YOU LOSE IF: (not before game tick 7401 OR NOT (lose ≤2 of your own units) OR your whole force is destroyed).
You have at most 84 decision turns; acting decisively and early matters.

Strict ordered chain: first reach staging point A (mid-west), then transit point B (north-centre), then seize the citadel C (far east) — IN THAT ORDER. Reaching C without having passed A then B counts for nothing. The strike is timed: C only counts after the hold. A greedy beeline to C, or idling, loses on the clock.
WIN WHEN: reach these waypoints IN ORDER (no skipping, no idling): region (35,20) r=3 → then → region (70,12) r=3 → then → region (110,20) r=3 AND not before game tick 1100 AND before game tick 2300.
YOU LOSE IF: (not before game tick 2301 OR your whole force is destroyed).
You have at most 30 decision turns; acting decisively and early matters.

Same ordered chain (A → B → C) but a longer prerequisite hold and a tighter terminal deadline — the strike window is narrow, so the order AND the timing both bind.
WIN WHEN: reach these waypoints IN ORDER (no skipping, no idling): region (35,20) r=3 → then → region (70,12) r=3 → then → region (110,20) r=3 AND not before game tick 1300 AND before game tick 2300.
YOU LOSE IF: (not before game tick 2301 OR your whole force is destroyed).
You have at most 30 decision turns; acting decisively and early matters.

Four-point ordered chain from a seed-chosen latitude, narrow timed window, and a loss cap — seize the citadel with the force near-intact, in order, inside the band.
WIN WHEN: reach these waypoints IN ORDER (no skipping, no idling): region (30,20) r=3 → then → region (60,30) r=3 → then → region (90,10) r=3 → then → region (110,20) r=3 AND not before game tick 1400 AND before game tick 2300 AND lose ≤1 of your own units.
YOU LOSE IF: (not before game tick 2301 OR NOT (lose ≤1 of your own units) OR your whole force is destroyed).
You have at most 32 decision turns; acting decisively and early matters.

No power is pre-placed. Build a power plant first, THEN the power-dependent barracks (tent) — a power-less barracks never completes. Win = own a 'tent' AND ≥3 buildings total. Building the barracks first never completes (no power); spamming power plants never gives a barracks; idling loses on the clock.
WIN WHEN: own ≥3 buildings total AND own a 'tent' AND before game tick 2400.
YOU LOSE IF: NOT (before game tick 2400).
You have at most 30 decision turns; acting decisively and early matters.

Build a defensive line to the EAST: stand up power, then the barracks it enables, then place at least two pillboxes (pbox — they need the barracks, which needs power) INSIDE the designated eastern region near (40,20). Skipping the chain never yields a pillbox; placing the pillboxes near the base (wrong direction) does not count; idling loses on the clock.
WIN WHEN: have 2 building(s) near (40,20) AND before game tick 3400.
YOU LOSE IF: NOT (before game tick 3400).
You have at most 40 decision turns; acting decisively and early matters.

Found the defensive line FAR to the east, near (60,20) — beyond the construction yard's reach. Creep the base east (each new power plant extends where you may build) until you can stand up the barracks and place at least two pillboxes (pbox → tent → power chain) inside the far region, before the tight deadline. Building near the spawn never reaches the region; a power creep that skips the barracks never yields a pillbox; idling loses.
WIN WHEN: have 2 building(s) near (60,20) AND before game tick 5800.
YOU LOSE IF: NOT (before game tick 5800).
You have at most 70 decision turns; acting decisively and early matters.

A single concentrated wave of 4 rifle infantry charges your base (fact+tent+powr already standing). Win by holding the line and sustaining production: build at least one pillbox (pbox) covering the lane mouth AND have ≥3 of your own units alive AND destroy ≥4 of the rushers before tick 4800. Stalling, building only pillboxes (no units), building no pillbox, or spending only on defenders without a pillbox all fail.
WIN WHEN: own ≥1 'pbox' building(s) AND keep ≥3 units alive AND destroy ≥4 enemy units AND before game tick 4800.
YOU LOSE IF: (not before game tick 4801 OR NOT (own ≥1 'fact' building(s))).
You have at most 60 decision turns; acting decisively and early matters.

The full rush band — 6 rifle infantry + 2 rocket soldiers — charges your base in one concentrated wave. Hold the line and keep producing: at least one pillbox at the lane mouth AND ≥3 own units AND ≥6 enemy kills AND ≥4 own buildings (the pre-placed fact+tent+powr plus at least one more — the pillbox counts) before tick 4800. The same lazy/single-axis policies that fail easy fail here.
WIN WHEN: own ≥1 'pbox' building(s) AND keep ≥3 units alive AND destroy ≥6 enemy units AND own ≥4 buildings total AND before game tick 4800.
YOU LOSE IF: (not before game tick 4801 OR NOT (own ≥1 'fact' building(s))).
You have at most 60 decision turns; acting decisively and early matters.

Two staggered rush bands hit your base from two different latitudes (north and south). The agent base stages from a seed-chosen latitude (the rush lane that matters varies by seed, so a single memorised pillbox cell cannot generalise). Hold the line, keep producing, and bleed less than four of your own: at least one pillbox at the lane mouth AND ≥3 own units AND ≥8 enemy kills AND ≥4 own buildings AND ≤3 own units lost before tick 4800. Stall, defence-only, economy-only, brute-army-only, and over-attrition all fail.
WIN WHEN: own ≥1 'pbox' building(s) AND keep ≥3 units alive AND destroy ≥8 enemy units AND own ≥4 buildings total AND lose ≤3 of your own units AND before game tick 4800.
YOU LOSE IF: (not before game tick 4801 OR NOT (own ≥1 'fact' building(s)) OR NOT (lose ≤3 of your own units)).
You have at most 60 decision turns; acting decisively and early matters.

Allocate the budget to a structural upgrade PLUS a quality force: build and place a second power plant (powr) and train at least 4 anti-vehicle infantry (e3). Spamming the cheap rifle (e1) clears neither bar. Generous clock; the only failure is an unfocused or late spend.
WIN WHEN: own ≥2 'powr' building(s) AND have ≥4 'e3' AND before game tick 2703.
YOU LOSE IF: not before game tick 2703.
You have at most 30 decision turns; acting decisively and early matters.

Tighter budget and clock. The same allocation, scaled: a second power plant plus 5 anti-vehicle infantry. The budget covers exactly this one coherent build — cheap-unit spam or an over-investment misses the deadline.
WIN WHEN: own ≥2 'powr' building(s) AND have ≥5 'e3' AND keep ≥5 units alive AND before game tick 2523.
YOU LOSE IF: not before game tick 2523.
You have at most 28 decision turns; acting decisively and early matters.

Lean budget, short clock, zero-attrition: build and place a second power plant and train 6 anti-vehicle infantry, losing nothing. Every credit and tick must convert to fielded force in the correct order; spam or an over-spend fails the clock.
WIN WHEN: own ≥2 'powr' building(s) AND have ≥6 'e3' AND lose ≤0 of your own units AND before game tick 2163.
YOU LOSE IF: (not before game tick 2163 OR NOT (lose ≤0 of your own units)).
You have at most 24 decision turns; acting decisively and early matters.

Loose clock. A refinery and one harvester are already placed next to two ore patches; just commit to harvesting promptly. The starting cash (300) is well below the bar (1500), so idling cannot satisfy it — only the harvest income closes the gap. The intended scripted harvester wins by ~tick 1350.
WIN WHEN: reach economy value ≥1500 (cash+ore) AND before game tick 1800.
YOU LOSE IF: (not before game tick 1801 OR your whole force is destroyed).
You have at most 22 decision turns; acting decisively and early matters.

Tighter clock and a higher bar (5500 economy value by tick 3000). Starting cash (1500) is well below the bar, and a single harvester at this map only produces ~3150 ore over 3000 ticks (total ~4650) — that misses the bar. A war factory and a construction yard are pre-placed: the intended play is to build one extra harvester (cost 1100) while the first keeps harvesting, then run both. Spending the budget on infantry / other units starves the income channel and loses; stalling loses.
WIN WHEN: reach economy value ≥5500 (cash+ore) AND before game tick 3000.
YOU LOSE IF: (not before game tick 3001 OR your whole force is destroyed).
You have at most 35 decision turns; acting decisively and early matters.

High bar (7500) and a short clock (3600 ticks). Starting cash (1200) is well below the bar; a single pre-placed harvester only reaches ~3200-5200 ore over 3600 ticks — that misses. The intended play is to build TWO extra harvesters (3 total) and feed all three ore patches in parallel. The starting position varies by seed (two spawn_point groups: NW or SW base) so a memorised single fixed plan can't generalise — the model must locate the pre-placed patches relative to its base each episode.
WIN WHEN: reach economy value ≥7500 (cash+ore) AND before game tick 3600.
YOU LOSE IF: (not before game tick 3601 OR your whole force is destroyed).
You have at most 42 decision turns; acting decisively and early matters.

WIDE path. The budget is exactly the cost of a second refinery plus its supporting power plant (1400 + 300 = 1700) and nothing more. Build and place a 2nd refinery and a 2nd power plant. Any credit spent on units instead leaves too little to finish both structures — a split budget builds neither in time.
WIN WHEN: own ≥2 'proc' building(s) AND own ≥2 'powr' building(s) AND before game tick 1893.
YOU LOSE IF: not before game tick 1893.
You have at most 24 decision turns; acting decisively and early matters.

WIDE path under a tighter clock, plus a small utilising force. The budget is exactly the wide expansion (1400 + 300) plus three infantry (3 × 100 = 2000) and nothing more. Build the 2nd refinery, 2nd power plant, AND three units. A fourth unit (or any other diversion) overspends and the second refinery never completes in time.
WIN WHEN: own ≥2 'proc' building(s) AND own ≥2 'powr' building(s) AND keep ≥3 units alive AND before game tick 2073.
YOU LOSE IF: not before game tick 2073.
You have at most 26 decision turns; acting decisively and early matters.

DEEP path. The budget converts entirely into a forward force: keep the single refinery and field 22 infantry (22 × 100 = 2200, the whole budget). Building a second refinery or power plant consumes credits the force needs and the bar is missed. Newly produced units must be moved clear of the barracks or production stalls.
WIN WHEN: keep ≥22 units alive AND own ≥1 'proc' building(s) AND before game tick 2613.
YOU LOSE IF: not before game tick 2613.
You have at most 32 decision turns; acting decisively and early matters.

Generous clock (40k ticks), generous coverage bar. Chain the four phases in order: explore >=35% of the map, build a refinery (proc), stand up >=4 combat units, and put at least one unit on the enemy base at (115, 32). Missing any of the four at episode end is a loss, not a draw.
WIN WHEN: reveal ≥35% of the map AND own a 'proc' AND keep ≥4 units alive AND get a unit into region (115,32) r=8 AND before game tick 10803.
YOU LOSE IF: not before game tick 10803.
You have at most 120 decision turns; acting decisively and early matters.

Tighter clock (32k ticks), coverage 45%, force >=5, refinery required, contact the enemy base. Same chain; less slack.
WIN WHEN: reveal ≥45% of the map AND own a 'proc' AND keep ≥5 units alive AND get a unit into region (115,32) r=7 AND before game tick 12603.
YOU LOSE IF: not before game tick 12603.
You have at most 140 decision turns; acting decisively and early matters.

Short clock (24k ticks), coverage 55%, force >=6, refinery required, both an enemy building discovered AND a unit on the enemy base. An attrition cap prevents brute-force loss-then-win. Skipping the scouting phase to rush tech-then-strike fails the coverage clause; skipping tech to rush units fails the has_building clause.
WIN WHEN: reveal ≥55% of the map AND spot ≥1 enemy buildings AND own a 'proc' AND keep ≥6 units alive AND get a unit into region (115,32) r=6 AND before game tick 14403 AND lose ≤4 of your own units.
YOU LOSE IF: (not before game tick 14403 OR your whole force is destroyed).
You have at most 160 decision turns; acting decisively and early matters.

Two candidate regions, one decoy. The objective marker (a building) sits in the far-NE corner; a decoy enemy *unit* cluster sits in the far-SE corner. Only the NE building counts toward the win. A single decisive trip straight to the NE corner finishes well inside the deadline; detouring to the SE decoy first, or sweeping both regions, overruns the clock and LOSES. Perception is honest — the only task is to commit to the correct region.
WIN WHEN: spot ≥1 enemy buildings AND before game tick 950.
YOU LOSE IF: (not before game tick 951 OR your whole force is destroyed).
You have at most 14 decision turns; acting decisively and early matters.

Three candidate regions, two decoys, a tighter deadline. Decoy enemy *unit* clusters sit in the far-SE corner and the mid-south edge; the real survivor marker is the building in the far-NE corner. From the start corner all three look equally plausible, but only the NE building counts. The deadline now permits exactly one decisive trip to the correct corner — committing to either decoy first, or sweeping, overruns it and LOSES. The model must reason about which region to commit to, not just navigate.
WIN WHEN: spot ≥1 enemy buildings AND before game tick 880.
YOU LOSE IF: (not before game tick 881 OR your whole force is destroyed).
You have at most 14 decision turns; acting decisively and early matters.

Three candidate regions, two decoys, a tight deadline, and an attrition constraint. One decoy in the far-SE corner is a HOSTILE cluster (stance 2) — a careless route into it gets the lone scout killed; a passive decoy sits on the mid-south edge. The real survivor marker is the building in the far-NE corner. The model must pick the commitment that (a) finds the building, (b) fits the clock — a detour to a decoy region first overruns it — and (c) avoids the lethal frontier, all without splitting (only one unit). The scout starts at a seed-chosen corner so the commitment decision must generalise across start positions.
WIN WHEN: spot ≥1 enemy buildings AND before game tick 1050 AND lose ≤0 of your own units.
YOU LOSE IF: (not before game tick 1051 OR NOT (lose ≤0 of your own units) OR your whole force is destroyed).
You have at most 15 decision turns; acting decisively and early matters.

A leashed rocket-infantry wall blocks the direct lane east to the objective and shreds anything that drives into the mid-band. Get the scout to the objective (far east, marked by the enemy structure) with ZERO losses, before the deadline. Driving straight east loses the scout in a turn; go the long way — climb to the open top edge (y≤3) or drop to the open bottom edge (y≥37), run east past the wall's tip, then turn in. Idling never commits and loses on the clock.
WIN WHEN: get a unit into region (120,20) r=7 AND lose ≤0 of your own units AND before game tick 2700.
YOU LOSE IF: (not before game tick 2701 OR NOT (lose ≤0 of your own units) OR your whole force is destroyed).
You have at most 32 decision turns; acting decisively and early matters.

The bottom edge is now also guarded — only the longer top-edge route is safe. The model must reason that the short lane is lethal, the bottom detour is also lethal, and the top detour (y≤3), though longer, is the only plan that satisfies both the zero-loss constraint and the tighter deadline.
WIN WHEN: get a unit into region (120,20) r=7 AND lose ≤0 of your own units AND before game tick 3100.
YOU LOSE IF: (not before game tick 3101 OR NOT (lose ≤0 of your own units) OR your whole force is destroyed).
You have at most 36 decision turns; acting decisively and early matters.

Both edges are contested: the bottom is fully blocked (lethal cluster) and the top edge has a defended pocket near x=100. The only winning plan threads the narrow safe seam — hug the top edge (y≤3) east past the central wall, then dip down to y≈10 to skirt BELOW the top pocket while staying clear of the (now distant) central wall, then turn in to the objective before the tight deadline. Risk and route must be traded against the clock, not avoided outright.
WIN WHEN: get a unit into region (122,20) r=6 AND lose ≤0 of your own units AND before game tick 3300.
YOU LOSE IF: (not before game tick 3301 OR NOT (lose ≤0 of your own units) OR your whole force is destroyed).
You have at most 40 decision turns; acting decisively and early matters.

Risk Dilemma \u2014 destroy the enemy's key economic buildings (construction yard + refinery). A heavy defensive cluster (two pillboxes, a Tesla coil and four rifles) blocks the DIRECT west-to-east lane; charging straight through it gets the force wiped before it ever reaches the base. The NORTHERN flank past that cluster is lightly held \u2014 take the safer route, reach the base, eliminate fact+proc before the clock runs out.
WIN WHEN: destroy the enemy fact+proc AND before game tick 4500.
YOU LOSE IF: (NOT (before game tick 4500) OR your whole force is destroyed).
You have at most 55 decision turns; acting decisively and early matters.

Same objective (fact+proc) and the same lethal central cluster, but a tighter clock AND an attrition cap: a costly brawl no longer counts as a win even if you reach the base. Take the safer northern flank \u2014 losing more than 7 units fails the raid.
WIN WHEN: destroy the enemy fact+proc AND before game tick 3600 AND lose ≤7 of your own units.
YOU LOSE IF: (NOT (before game tick 3600) OR NOT (lose ≤7 of your own units) OR your whole force is destroyed).
You have at most 45 decision turns; acting decisively and early matters.

fact+proc destroyed, tight clock, strict attrition: only a clean raid that takes the lightly-held northern flank and avoids the central pillbox/Tesla cluster wins; brute-forcing the direct lane (or dithering) loses. Lose more than 5 units and the raid fails.
WIN WHEN: destroy the enemy fact+proc AND before game tick 2700 AND lose ≤5 of your own units.
YOU LOSE IF: (NOT (before game tick 2700) OR NOT (lose ≤5 of your own units) OR your whole force is destroyed).
You have at most 35 decision turns; acting decisively and early matters.

The Gauntlet — run the single defended corridor and destroy the enemy's key economic buildings (construction yard + refinery). A forward picket screens a base ring; a careless all-in charge is ground down before reaching fact+proc, and chasing every stray infantryman bleeds the column dry. Sequence the run: clear the gate, then drive the surviving force onto the economy with a sustainable attrition.
WIN WHEN: destroy the enemy fact+proc AND before game tick 8000 AND lose ≤12 of your own units.
YOU LOSE IF: (not before game tick 8001 OR your whole force is destroyed OR NOT (lose ≤12 of your own units)).
You have at most 90 decision turns; acting decisively and early matters.

Same single corridor and objective (fact+proc) and clock as easy; the new axis is a tighter attrition cap — a costly all-in run no longer counts as a win, the picket must be cleared cleanly.
WIN WHEN: destroy the enemy fact+proc AND before game tick 8000 AND lose ≤9 of your own units.
YOU LOSE IF: (not before game tick 8001 OR your whole force is destroyed OR NOT (lose ≤9 of your own units)).
You have at most 90 decision turns; acting decisively and early matters.

Same single corridor, same fact+proc objective, same attrition cap as medium; the new axis is a tight clock — the gauntlet must be cleared and the economy levelled fast, not at leisure.
WIN WHEN: destroy the enemy fact+proc AND before game tick 4500 AND lose ≤9 of your own units.
YOU LOSE IF: (not before game tick 4501 OR your whole force is destroyed OR NOT (lose ≤9 of your own units)).
You have at most 55 decision turns; acting decisively and early matters.

Three mutually exclusive paths to victory. Pick ONE and commit the whole budget to it:
  EXPAND — build+place a 2nd refinery (proc) at (60,20) in the
           far region (cost 1400).
  TECH   — build+place a war factory (weap, cost 2000).
  ARMY   — train ≥12 e1 infantry (cost 1200) AND defeat at least
           1 enemy unit (the hunt harasser).
Easy budget gives small slack (3000 credits vs the binding 2400) and a generous clock so the intended path completes comfortably — but the budget still does not fund any TWO paths together (1400+2000=3400, 1400+1200=2600, 2000+1200=3200), so a hedged "do a bit of each" spend fails every arm.
WIN WHEN: (have 1 building(s) near (60,20) AND before game tick 5400 OR own a 'weap' AND before game tick 5400 OR keep ≥12 units alive AND destroy ≥1 enemy units AND before game tick 5400).
YOU LOSE IF: (not before game tick 5401 OR NOT (own a 'fact')).
You have at most 60 decision turns; acting decisively and early matters.

Same three arms (EXPAND proc at (60,20) / TECH weap / ARMY ≥12 e1 + 1 kill), now under a BINDING budget: 2400 credits funds EXACTLY ONE full path. Any hedged spend (proc + a few units, weap + a few units, etc.) leaves no arm complete before the shorter deadline.
WIN WHEN: (have 1 building(s) near (60,20) AND before game tick 4500 OR own a 'weap' AND before game tick 4500 OR keep ≥12 units alive AND destroy ≥1 enemy units AND before game tick 4500).
YOU LOSE IF: (not before game tick 4501 OR NOT (own a 'fact')).
You have at most 50 decision turns; acting decisively and early matters.

Tightest version of the trilemma. Budget cut to 2200 (TECH eats 91% of it; EXPAND 64%; ARMY 55%), the clock is shorter, AND an attrition cap is added: lose more than 3 units and the raid fails. The hunt harasser is doubled, so the ARMY arm must actually defend its force while completing 12 units + a kill. Two agent spawn_point groups vary the harvester staging by seed (the round-robin start the held-out-seed metric rewards).
WIN WHEN: (have 1 building(s) near (60,20) AND before game tick 3600 AND lose ≤3 of your own units OR own a 'weap' AND before game tick 3600 AND lose ≤3 of your own units OR keep ≥12 units alive AND destroy ≥1 enemy units AND before game tick 3600 AND lose ≤3 of your own units).
YOU LOSE IF: (not before game tick 3601 OR NOT (own a 'fact') OR NOT (lose ≤3 of your own units)).
You have at most 40 decision turns; acting decisively and early matters.

Simple Tech Chain — fact, powr, and proc are pre-placed; build War Factory → Service Depot, then train one Heavy Tank. The decision is recognising 3tnk needs weap + fix in that order; the budget covers the chain with modest slack. Tests the agent can sequence a deep enough prereq chain (proc→weap→fix→3tnk) without attempting to train 3tnk before fix exists.
WIN WHEN: have ≥1 '3tnk' AND own ≥1 'fix' building(s) AND before game tick 5403.
YOU LOSE IF: not before game tick 5403.
You have at most 60 decision turns; acting decisively and early matters.

Parallel Dependency Optimization — only fact + powr pre-placed. Produce one Heavy Tank PLUS one Rocket Soldier PLUS three Rifle Infantry, with power_surplus ≥ 0 at the deadline. The combined build (proc + barr + weap + fix = -110 power) drains net power below zero against a 100-power base, so a SECOND power plant is mandatory. Tight clock forces queueing the infantry branch (barr) in parallel with the vehicle branch (weap+fix); serial production blows the clock on idle time.
WIN WHEN: have ≥1 '3tnk' AND have ≥1 'e3' AND have ≥3 'e1' AND keep power surplus ≥0 AND before game tick 7203.
YOU LOSE IF: not before game tick 7203.
You have at most 80 decision turns; acting decisively and early matters.

Adaptive Tech Transition — mid-game start: fact + powr + proc + barr and three Rifle Infantry are already on the field (an infantry build committed). Three enemy Heavy Tanks sit passive on the east side, hidden by fog. The agent must scout, recognise that infantry is the wrong counter to heavy armor, and pivot production to FIVE Rocket Soldiers (e3, anti-armor) before the deadline. Overproducing infantry past ten = fail (doubling down on the wrong plan exhausts the budget). The transition — cancel planned infantry, fund anti-armor, scout — is the test.
WIN WHEN: spot ≥2 enemy units AND have ≥5 'e3' AND before game tick 8103.
YOU LOSE IF: (not before game tick 8103 OR have ≥10 'e1').
You have at most 90 decision turns; acting decisively and early matters.

Pulsed strike: hit Cluster A first, then PULL BACK to the lull safe-point (south of the through-route) while the lull window passes, then push on and hit Cluster B. Order matters: the win only counts if you visit A → lull safe-point → B in that order. Continuous attack through the centre skips the lull and never wins. You need ≥7 kills across both windows and ≤4 own losses. Window 2 cannot open before tick 3000.
WIN WHEN: reach these waypoints IN ORDER (no skipping, no idling): region (35,20) r=3 → then → region (55,36) r=3 → then → region (105,20) r=3 AND not before game tick 3000 AND destroy ≥7 enemy units AND lose ≤4 of your own units AND before game tick 4500.
YOU LOSE IF: (not before game tick 4501 OR NOT (lose ≤4 of your own units) OR fewer than 2 of your units remain).
You have at most 52 decision turns; acting decisively and early matters.

Same pulsed cycle (A → lull safe-point → B) but the through- route now has a Tesla coil (tsla, anti-everything) at the narrow point. Trying to keep firing between the clusters runs your column past the Tesla and burns through the loss cap. Pull back south to the lull safe-point during the lull window and stay out of the Tesla's arc; push on to Cluster B only after the lull opens. ≥7 kills, ≤4 losses, in budget.
WIN WHEN: reach these waypoints IN ORDER (no skipping, no idling): region (35,20) r=3 → then → region (55,36) r=3 → then → region (105,20) r=3 AND not before game tick 3000 AND destroy ≥7 enemy units AND lose ≤4 of your own units AND before game tick 4500.
YOU LOSE IF: (not before game tick 4501 OR NOT (lose ≤4 of your own units) OR fewer than 2 of your units remain).
You have at most 52 decision turns; acting decisively and early matters.

Two Tesla coils now bracket the through-route — the lethal corridor is wider, and any attempt to push through it during the lull will shred the column. You may also start from the NORTH or the SOUTH staging (seed-chosen). Strike Cluster A, pull back to the lull safe-point south of the corridor, wait out the lull, then advance to Cluster B. ≥7 kills, ≤2 losses, in budget.
WIN WHEN: reach these waypoints IN ORDER (no skipping, no idling): region (35,20) r=3 → then → region (55,36) r=3 → then → region (105,20) r=3 AND not before game tick 3000 AND destroy ≥7 enemy units AND lose ≤2 of your own units AND before game tick 4500.
YOU LOSE IF: (not before game tick 4501 OR NOT (lose ≤2 of your own units) OR fewer than 2 of your units remain).
You have at most 52 decision turns; acting decisively and early matters.