SRD 'Immediate Actions': 'You also cannot use an immediate action if you are flat-footed.' _check_reactions now skips states carrying the flat_footed condition before consulting the reaction policy. The engine never applies flat_footed itself (latent gate), so the unit test injects the condition via active_conditions, with a control test proving the same setup fires when not flat-footed.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
SRD 'Ready - Initiative Consequences of Readying': 'If you come to your next action and have not yet performed your readied action, you don't get to take the readied action (though you can ready the same action again).' _take_turn's start-of-turn block now clears state.readied/readied_trigger (logged 'ready expired'), before the policy call; the ready still survives the round boundary and can fire in the next round as long as the readier's own turn has not arrived. Updated the two tests that pinned the old infinite persistence and added cross-round fire + expiry tests.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
CRB 'Cast a Spell - Touch Attacks': 'You can score critical hits with either type of attack as long as the spell deals damage'; CRB 'Range - Touch': a touch spell threatens on a natural 20 and deals double damage on a successful confirmation. The melee and ranged spell-touch gates now add the target's cover bonus to touch AC (new _attack_cover_bonus helper shared with resolve_attack), roll the concealment miss chance after the hit (blinded attacker = total concealment convention), and resolve crit threat/confirm with damage doubling for damage-dealing touch spells. Non-damage touch spells do not threaten (no die consumed). Golden transcript unaffected.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
CRB 'Conditions - Blinded': 'All opponents are considered to have total concealment (50% miss chance) against the blinded character.' The sneak-attack gate and the AoO suppression gate now apply the same max(target_concealment, 50) convention as _resolve_hit_damage: a blinded attacker cannot sneak attack and its AoOs are suppressed without spending the budget.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
SRD 'Attack - Shooting or Throwing into a Melee': a target at least 10 ft from the nearest friendly character lets you avoid the -4 penalty entirely (0, not -2). The -2 is reserved for targets two size categories larger than the engaged friendlies (0 at three+), which is not modeled (documented). Golden transcript unaffected (shot still misses).
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
Two documented limitations, no mechanical change: (a) Magic Missile fires a single 1d4+1 missile - the SRD scaling to up to five missiles per caster level and multi-target splitting is not modeled; (b) the engine applies spell resistance to every targeted spell, so the few spells whose SRD entry says "SR: no" (e.g. Acid Arrow) are not distinguished - a per-spell SR flag was deliberately not added.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (SRD spells/fear.md): creatures become panicked unless they succeed on a Will save - the JSON applied frightened instead. Condition fixed to panicked. The shaken-for-1-round on a successful save needs partial-save support for conditions and the 30-ft cone needs a cone template, neither of which the engine has; both documented in README as out of scope rather than built.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (SRD spells/acid-arrow.md): "Range long (400 ft. + 40 ft./level)" - the JSON declared medium. Fixed to long. The spell's ongoing acid (2d4 per extra round, one per three caster levels, unless neutralized) is not modeled: the engine has no per-spell damage-over-time subsystem, documented in README rather than built.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (SRD spells/burning-hands.md): "1d4 points of fire damage per caster level (maximum 5d4)" — the JSON encoded 1d4+1, a flat bonus the spell never grants. Base is now 1d4. Per-caster-level scaling and the 15-ft. cone are not modeled by the engine (no CL-scaled dice, no cone template); documented in README rather than built.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (APG Cackle): "(Su) Free Action; ... extends the duration of all evil eye, fortune or misfortune hexes on all creatures within 30 feet by 1 round." The engine spent a hex use, acted only on the resolved target, and extended every hex-sourced condition (including slumber, which RAW cackle does not touch). Cackle is now dispatched before the resource/target gate (no cost), occupies the FREE budget slot via _effective_slot (the hex kind stays standard for evil_eye/slumber), and rebuilds the evil_eye buff list of every creature within 30 ft in a single bounded pass. Fortune/misfortune are not implemented, so evil_eye is the only extendable family member — documented in README. test_cackle.py rewritten for the new semantics (extension, range, free slot, no cost).
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (APG Slumber): the target "remains asleep for a number of rounds equal to the witch's level" (Will negates). _hex_slumber hardcoded 1 round; it now uses the caster's level. The RAW clause that the hex ends when the target takes damage is not modeled (no wake-on-damage hook in the engine) — documented in README. Two behavioral tests pin the failed-save duration (= level) and the saved case.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (APG Evil Eye): the target takes -2 on ONE chosen option (AC, ability checks, attack rolls, saving throws or skill checks) for 3 rounds, and "A Will save reduces this to just 1 round" — it never negates. _hex_evil_eye applied shaken (-2 attack AND saves) for 1 round with the save negating entirely. It now adds a targeted -2 attack ActiveBuff for 3 rounds, or 1 round on a successful Will save. Simplification documented in README: the option choice is not modeled (attack rolls are the engine's target). The hex-DC behavioral test now reads the -2 attack (3r) transcript line.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (SRD spells/bleed.md): "Saving Throw Will negates" and the target takes 1 point of damage per round. Two divergences: the JSON declared a Fortitude save, and the bleeding condition never set the engine's bleed_damage field, so the spell was a no-op. Bleed now saves Will; _apply_condition_effect sets bleed_damage = 1 when a spell applies bleeding, and _decrement_durations lands the final tick as the last bleeding condition expires before clearing the field (heal-path clearing already existed in _stop_bleeding). README updated.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (SRD spells/sound-burst.md): "Every creature in the area takes 1d8 points of sonic damage and must succeed on a Fortitude save to avoid being stunned for 1 round" — the Fortitude partial negates the stun only, never the dice. The damage component declared half_on_save: true, halving the sonic damage on a successful save. DamageEffect gains a save_negates flag (default True, existing behavior unchanged); sound_burst.json declares half_on_save: false + save_negates: false so the 1d8 is full regardless while the condition keeps save_negates. README updated.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (CRB Movement): "You can't run or charge across difficult
terrain". _execute_run checked the explicit destination against the full
cost field (which routes through difficult terrain at 2x) but descended
with avoid_difficult=True, so a destination beyond rubble was accepted
(ok=True) yet silently truncated at the obstacle. _destination_reachable
now takes the same avoid_difficult constraint (difficult cells removed
from the field; a destination whose footprint touches difficult terrain is
not a legal run endpoint either), so verdict and path agree. The test that
pinned the old silent-stop-short convention now expects
unreachable_destination; README updated.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (APG Reposition / Drag): "For every 5 by which your attack exceeds
your opponent's CMD, you can move/drag the target an additional 5 feet."
Both handlers now walk 1 + margin//5 cells along the chosen (or auto-picked)
direction, stopping at the first obstacle or occupied square, mirroring the
bull rush scaling. README updated.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (APG Dirty Trick): "This condition lasts for 1 round. For every 5
by which your attack exceeds your opponent's CMD, the penalty lasts 1
additional round" — rounds_left = 1 + margin//5. The RAW condition list
(blinded, dazzled, deafened, entangled, shaken, sickened) does not include
nauseated, so it is removed from TrickCondition. README updated.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (SRD Combat Maneuvers): "Any penalties to a creature's AC also
apply to its CMD." _maneuver_cmd now folds negative AC modifiers into the
sum regardless of bonus type (grappled -2, other untyped penalties), and —
a maneuver being a melee attack roll — the prone -4 melee AC penalty.
Positive modifiers keep the eight SRD-listed types. The test that pinned
the old ignore-penalties convention is updated; README documents the rule.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (CRB Critical Hits): "additional damage dice from weapon special
abilities (such as flaming) are not multiplied when you score a critical
hit". _resolve_hit_damage now multiplies only the physical dice plus flat
bonuses by crit_mult; energy-typed component dice are added at 1x (the
energy share on the DR path and the per-component resistance step drop
their crit factor). Sneak precision dice were already excluded. README
and the module convention list updated.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (APG Steal): "Items held in the hands (such as wielded weapons or
wands) also cannot be taken with the steal maneuver—you must use the disarm
combat maneuver instead." The model carries no non-held items, so a
successful steal now takes nothing and logs
"nothing stealable — wielded weapons require disarm". README updated; the
retrieve_weapon tests keep locking the restoration mechanism directly.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (APG Reposition / Drag): "You can only reposition/drag an opponent
that is no more than one size category larger than you" — the same cap the
SRD puts on trip/bull rush/overrun. reposition and drag join
_SIZE_RESTRICTED_MANEUVERS, so a two-categories-larger target is rejected
outright (no roll, no AoO). README updated.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
Record the three limitations deliberately left out of the L1-L10 conformance pass in the README: no surprise round (the combatant model has no awareness field), one_round casting clamped to a full-round action (no pending-cast state across turns), and Tiny/Diminutive/Fine reach and co-occupancy abstracted to the 5-ft grid.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (SRD Immediate Actions): "Using an immediate action on your turn is the same as using a swift action and counts as your swift action for that turn." The engine now tracks the acting combatant (_turn_actor); an immediate fired during its OWN turn sets swift_used_this_turn instead of immediate_used, so swift actions later in the turn's sequence are dropped (downgraded:immediate_swift) and no second immediate fires that round, while the next turn's swift stays free. Off-turn immediates keep consuming the next swift. The off-turn consumption test was retargeted to a true off-turn event and an in-turn test added; README updated (Actions immediates + budget classification).
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (SRD Conditions, nauseated): "The only action such a character can take is a single move actions per turn." _NAUSEA_MOVE_KINDS now holds only genuine move actions (move, stand_up, retrieve_weapon, repair_weapon): the 5-foot step and dropping prone are free actions and are dropped by the nauseated filter. The pinned filter test was retargeted; README updated.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (SRD Trip / Bull Rush / Overrun): "You can only [affect] an opponent who is no more than one size category larger than you." _execute_maneuver now rejects trip, bull_rush and overrun against a target two or more size categories larger: no roll, no provoked AoO, logged rejection. A _SIZE_ORDER ranking (Fine..Colossal, unknown -> Medium) backs the check; other maneuvers keep no size cap. README updated.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (SRD Trip): "If your attack fails by 10 or more, you are knocked prone instead" - a trip whose total lands at CMD-10 or below now knocks the attacker prone (1 round). PF1e (SRD Overrun): "If your overrun attempt fails, you stop in the space directly in front of the opponent, or the nearest open space in front of the creature" - a failed overrun from beyond adjacency walks the attacker to the free approach cell nearest the target; an already-adjacent attacker stays put. README updated.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (SRD Bull Rush): "your target is pushed back 5 feet. For every 5 by which your attack exceeds your opponent's CMD you can push the target back an additional 5 feet." _bull_rush_push (and the explicit-direction branch) now walk 1 + margin//5 cells, stopping at the first obstacle; a natural-20 auto-success below CMD floors the margin at 0 (base push only). README updated.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (SRD Movement, Position and Distance): "You can move through a square occupied by a friendly character, unless you are charging." The move and run descents now route their Dijkstra cost field and greedy steps over hostile-only blocking (_blocked_hostile), so allies' squares are passable; a friendly square may only be entered when a free cell descends from it and the path tail is trimmed, so movement never ENDS on an occupied cell. Charge, withdraw, the auto 5-foot step and the Observation cost fields keep full blocking (documented). Corridor tests added; the Combat Reflexes multi-AoO transcript is unchanged. README updated.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (SRD Combat Maneuvers): "Rolling a natural 20 while attempting a combat maneuver is always a success (except when attempting to escape from bonds)" - escaping a grapple is a combat maneuver, not bonds. _maneuver_escape now auto-succeeds on a natural 20 below CMD (the natural 1 auto-failure is unchanged). README updated.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (SRD Measuring Distance): "You can't move diagonally past a corner (even by taking a 5-foot step)." _step5_to now rejects a diagonal destination whose two flanking squares are not both passable, reusing Grid.diagonal_allowed - the same corner rule the regular movement already applies. README updated.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (SRD Combat Maneuver Defense): "A creature can also add any circumstance, deflection, dodge, insight, luck, morale, profane, and sacred bonuses to AC to its CMD [...] A flat-footed creature does not add its Dexterity bonus to its CMD." _maneuver_cmd now sums the target's eligible AC bonus modifiers (plus the dodge/circumstance runtime bonuses from total defense, fight defensively and aid another) and drops the Dexterity bonus (keeping any penalty) while the target denies its Dexterity to AC. README updated.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (SRD Combat Maneuvers): "Creatures that are size Tiny or smaller use their Dexterity modifier in place of their Strength modifier to determine their CMB." _maneuver_cmb now swaps Str for Dex when the size is Tiny, Diminutive or Fine. True 0-reach and co-occupancy for Tiny-or-smaller stay unmodelled (5-ft grid abstraction), documented in the README.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (CRB rogue Sneak Attack): "A rogue cannot sneak attack while
striking a creature with total concealment." _sneak_attack_dice now gates
on miss_chance >= 50 instead of any concealment > 0, so a blurred (20%)
target can still be sneak attacked while a displaced (50%) one cannot. The
test that pinned the old any-concealment denial was retargeted to total
concealment and a partial-concealment case added; README updated.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (CRB Power Attack): "take a -1 penalty on all melee attack rolls ...
The bonus damage does not apply to touch attacks." A melee touch (kind
"touch", no range increment) is a melee attack, so _power_attack_amt now
rides its attack and confirmation rolls; the +2X/+3X damage bonus stays
gated on kind == "melee". Ranged attacks and rays are unaffected. The
test that pinned the old melee-weapon-only simplification was rewritten to
lock the corrected behavior; README and abilities.py updated.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (CRB Cover, Big Creatures and Cover): when making a melee attack
against a multi-square creature you can pick any of the squares it
occupies to determine cover. has_cover_cells/cover_bonus_cells now
aggregate the target side by the MINIMUM over target cells for melee and
keep the maximum (any line to any corner) for ranged; the attacker-side
minimum is unchanged. 1x1 geometry is bit-identical; no data case changes
today, locked by a unit test (Large target: melee 0, ranged +4).
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (CRB Cover): "You can't execute an attack of opportunity against an
opponent with cover relative to you"; (CRB Concealment): the same holds
against total concealment. _resolve_aoo now checks terrain cover
(cover_bonus_cells, melee line) and miss_chance >= 50 before striking,
logs the suppression and leaves the per-round budget unspent. Partial
concealment (blurred 20%) and normal attacks are unaffected.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (CRB Damage Reduction): "Damage reduction does not negate ...
energy damage dealt along with an attack." _resolve_hit_damage now
splits the post-crit total and applies DR only to the non-energy share
(physical dice + flat bonuses + precision dice); energy components are
reduced solely by the per-component energy resistance/immunity step.
DR-bypass semantics (multi-component "and" requires every component)
are unchanged. Pure-physical weapons keep bit-identical arithmetic and
the golden transcripts are untouched.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (CRB Combat Maneuvers): "If you are hit by the target, you take the
damage normally and apply that amount as a penalty to the attack roll to
perform the maneuver." Capture the provoked AoO's damage in
_execute_maneuver and subtract it from the check total (floored at 0;
natural 20 still succeeds, natural 1 still fails; dropping below 0 hp
still forfeits). Escape and the Improved-feat exemption are untouched.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (CRB Attacks of Opportunity): moving out of more than one square
threatened by the same opponent in the same round counts as one
opportunity for that opponent. Track per-round movement provocation in
_move_aoo_provoked (cleared with _aoo_used at round start); the stand-up
trigger passes movement=False and still provokes. Ranged, casting and
maneuver triggers are untouched.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e CRB 'Overrun': 'If your maneuver is successful, you move through
the target's space. If your attack exceeds your opponent's CMD by 5 or
more, ... the target is knocked prone.' The engine knocked the target
prone on ANY success. A plain success (margin 0-4) now only moves the
attacker through; prone requires total - CMD >= 5 (_OVERRUN_PRONE_MARGIN).
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e CRB 'Movement, Position and Distance - Difficult Terrain': 'You
can't run or charge across difficult terrain.' The run action reused the
move machinery (_descend_path), so it crossed rubble at 2x cost within
the 4x budget. The descent now takes an avoid_difficult option that
refuses every step whose footprint would touch a cell with move cost > 1
(the same rejection charge applies in _charge_path_clear); the run
moves up to the obstacle and stops there (deterministic, documented).
Move and charge behavior are unchanged.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e CRB 'Big and Little Creatures in Combat': a creature's natural
reach is a disk (a Large creature still threatens and strikes adjacent
squares), while a reach weapon doubles the wielder's natural reach but
'can't strike at their natural reach or less' (a ring whose dead zone is
the natural reach). The engine keyed the ring off reach_ft > 5, so
reach-10 natural attacks (giant belostomatid bite/claw, mounted
hooves) wrongly got a 5-ft dead zone.
AttackSpec gains reach_weapon (parsed from the combatant JSON); the
disk-vs-ring rule is centralized in _melee_reach_zone_cells and applied
consistently to weapon selection, melee usability, threatened squares
and charge end placement. Esha's Lance is the only reach weapon in the
shipped data; Large natural attackers keep the default (disk).
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (CRB, Disabled): a disabled creature can take move actions without further injuring itself. retrieve_weapon and repair_weapon are budget-classified move actions (_ACTION_ECONOMY_MOVE_KINDS) but were absent from _DISABLED_MOVE_KINDS, so a creature at 0 hp paid 1 hp to draw or mend a weapon. Added both kinds to the exempt set.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
A freshly-delayed combatant is re-inserted into the round order and _take_turn runs a second time in the same round. That second pass re-ran _decrement_durations and _turn_gate, double-ticking bleed, double-decrementing buff/condition durations and re-firing the fear aura. The start-of-turn block is now skipped when delayed_this_round is already set at entry (cleared at round start, set only by a delay action); the delayed turn still gets its policy call, budget and execution.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (CRB, Stable Characters and Recovery): characters taking continuous damage such as bleed automatically fail all Constitution checks made to stabilize, losing 1 hp per round in addition to the continuous damage. _process_dying never applied bleed (only _turn_gate, reached by active creatures did), so a bleeding dying creature lost 0 hp per round. It now ticks bleed_damage at the start of the dying turn and auto-fails the check (no d20 consumed), then evaluates death.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e (CRB, Staggered): a staggered creature takes a single move OR standard action but can still take free, swift, and immediate actions. _staggered_filter returned only the first non-full-round action, so a disabled combatant with a queued swift bonus (deed/inspire_courage) lost its standard action. It now keeps every swift and free action plus the first move-or-standard, preserving the full_attack -> single attack downgrade.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e CRB (Bardic Performance - Inspire Courage): the attack/weapon-damage
bonus is a COMPETENCE bonus (not morale), so it stacks with morale sources
like Heroism; the performance also grants a +1 MORALE bonus on saving throws
against charm and fear effects, implemented as two descriptor-conditional
will modifiers (fear / mind_affecting) that dedupe via morale stacking.
Simplifications (documented): flat +1 (no scaling per four bard levels) and
mind_affecting stands in for the missing charm descriptor.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e CRB (Magus Spell Combat): as a full-round action the magus casts a
standard-action spell and makes all his melee attacks at -2. The penalty is
threaded through the full-attack chain (_execute_attack -> _full_attack ->
_full_attack_weapon -> _attack/_resolve_swing -> resolve_attack bonus_override)
so it applies to every attack roll and crit confirmation. Simplification: the
arcane pool cost is still not modeled.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e CRB (Channel Energy): creatures damaged by channeled energy get a Will
save (DC 10 + half the channeler's level + her Charisma modifier) for half
damage (floor). Harm mode only; heal mode unchanged. Charisma is used RAW
even for casters whose spellcasting ability differs (channeling is a
Charisma class feature).
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
PF1e CRB (Magic - Ranged Touch Attacks): spells delivered by ray require a
ranged touch attack (BAB + Dex vs touch AC, no range increment) to hit a
hostile target; a miss wastes the slot. SpellSpec gains requires_ranged_touch,
set on the 8 SRD-confirmed rays (Ray of Frost, Ray of Enfeeblement, Scorching
Ray, Acid Arrow, Acid Splash, Disrupt Undead, Shard of Chaos, Spear of Purity).
Magic Missile and area spells are unaffected. Simplification (documented): no
size modifier / Weapon Finesse on the roll.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
ctan