Add mission customization, HQ support and aggression-driven AI
Integrate faction and random settings menus, generated objectives, scouts, convoys, air support, mortar stations and body-intel recovery. Add identified-body aggression, finite hunter UAZ deployment, dedicated objective defence and maximum AI skill with separate reduced aim accuracy. Include policy tests, native probes, evidence and documentation.
This commit is contained in:
@@ -0,0 +1,45 @@
|
||||
# Prison opening and intel debug command
|
||||
|
||||
The prison receives a finite 3–5-person squad drawn from the selected enemy
|
||||
faction's group profile. Selection happens before the native member spawn request.
|
||||
An RTO slot is preserved where available, and preparation confirms an actual
|
||||
large radio in the squad's inventory. For profiles without one, a soldier equips
|
||||
an AN/PRC-77 (US/USAF) or R-107M; the native loadout operation can replace an
|
||||
existing backpack. No further radio grants occur after initial preparation.
|
||||
The opening squad is placed once at separate random clear positions 12–24 m
|
||||
around the prison, while still frozen. Other sites retain their authored squad sizes. Existing prison AI freeze/release
|
||||
hooks still apply.
|
||||
|
||||
Administrators can use `/esc cheat intel` from their current living prisoner.
|
||||
The server checks six separate nearby ground positions, then creates six normal
|
||||
intel folders. If placement or creation fails, no partial batch is retained.
|
||||
The command does not insert anything directly into inventory or reveal locations;
|
||||
picking up a folder invokes the ordinary discovery rules. Seated characters and
|
||||
positions without enough accessible ground are refused.
|
||||
|
||||
## Verification, 22 September 2026
|
||||
|
||||
`python tools/check.py` passed after the production changes. The admin command
|
||||
regression includes authorized dispatch, unauthorized refusal, a placement-failure
|
||||
reply, and no inventory-grant side effects.
|
||||
|
||||
The `opening-tools` Core Workbench fixture completed with 22 assertions and no
|
||||
script/runtime failures in:
|
||||
`/tmp/erri-opening-tools2/logs/logs_2026-09-22_15-15-11`.
|
||||
It exercises roster sizes 3, 4 and 5; physical squad size; frozen living guards;
|
||||
a physically equipped manpack; native fallback equipment on a non-RTO; unauthorized
|
||||
command refusal; six fresh ground folders; actual pickup and normal discovery;
|
||||
and normal prison unlock. Source was unchanged throughout the run, dependencies
|
||||
were unchanged, and the runner confirmed cleanup. This is local Workbench
|
||||
integration, not a dedicated multiplayer test.
|
||||
|
||||
The strict native reporter still returns failure for existing RHS/WCS localization
|
||||
and world `Parent`/`SlidingTrackMaterial` resource errors. The same errors are in
|
||||
`/tmp/erri-mortar-station-fire2` from before these changes. They were not suppressed.
|
||||
|
||||
The live RKS #2 startup log for 22 September at 13:00 UTC records its extraction
|
||||
Mi-8 alive, grounded, stable and fueled around coordinates 7012/7500. The local
|
||||
fixture also confirms that aircraft is initially usable. Those authored coordinates were off the airfield. The follow-up change moves
|
||||
the extraction helicopter to a validated airfield pad and rejects obstructed or
|
||||
off-airfield placement; see [airfield and map changes](airfield-extraction-and-map-icons.md).
|
||||
The earlier log establishes initial aircraft condition, not the later crash sequence.
|
||||
Reference in New Issue
Block a user