Capture current mission, AI, map and administration development
Include the current addon, terrain integrations, tools, regression tests and recorded development evidence.
This commit is contained in:
@@ -0,0 +1,14 @@
|
||||
{
|
||||
"conclusion": "IsAIActivated does not identify prison hold. SDK explicitly says agents are disabled at maxLOD. The failed all-active assertion does not prove a prison-release defect. Individual failed agent LOD/cause was not logged, so distance dormancy remains a source-supported explanation, not established attribution.",
|
||||
"sdk": {
|
||||
"path": "/tmp/erri-vanilla-stock/scripts/Game/generated/AI/AIAgent.c",
|
||||
"sha256": "83d066ddd047136cc045228ce41e6a5f50f2b603c33fc7a0b4140061844f2a98",
|
||||
"lines": "20-37 GetLOD/SetPermanentLOD/PreventMaxLOD/AllowMaxLOD and native maxLOD disable semantics"
|
||||
},
|
||||
"phase_history": "Smoke EOnInit subscribes before scheduled StartScenario (1500ms); crash flag assigned before subscription. ERR_SmokePhase records any entered PRISON. Initial route requires RADIO_SEARCH. Existing source ERR_PrisonPhaseChanged unsubscribes and releases on current==RADIO_SEARCH without depending on previous; test now explicitly covers0->2 and1->2.",
|
||||
"limits": [
|
||||
"Probe enumerates currently retained force-ledger group agents; no claim of a historical roster snapshot or replacement detection.",
|
||||
"Active/dormant counts are diagnostic only and do not attribute dormancy to one engine subsystem.",
|
||||
"No native rerun by reviewer; no proof of crash full route yet."
|
||||
]
|
||||
}
|
||||
Reference in New Issue
Block a user