Improve paired cast continuity and wording

README.md

@@ -52,6 +52,12 @@ more natural language. Connect the prompt builder's `metadata_json` output to
 `prompt`; in that case the node falls back to prompt-label parsing or comma-tag
 cleanup.
 
+When connected to `SxCP Insta/OF Prompt Pair` metadata, the naturalizer emits a
+single combined natural caption with the shared descriptor plus separate
+softcore and hardcore version descriptions. It uses the final selected
+expression and composition from the generated rows, including any expression
+pool and intensity settings.
+
 Naturalizer controls:
 
 - `input_hint`: `auto`, `metadata_json`, or `caption_or_prompt`.
@@ -100,6 +106,12 @@ shared primary creator descriptor, then returns both a softcore prompt and a
 hardcore prompt from that same descriptor. This is useful when you want the same
 person/look/scene continuity but need two different prompt strengths.
 
+When the hardcore cast includes partners, pair mode also creates deterministic
+shared cast descriptors such as `woman A / primary creator` and `man A`. Use
+`softcore_cast=same_as_hardcore`, `hardcore_cast=couple`, and
+`continuity=same_creator_same_room` when you want a soft prompt with the same
+woman, man, and location as the hardcore prompt.
+
 It outputs:
 
 - `softcore_prompt`

caption_naturalizer.py

@@ -180,6 +180,18 @@ def _clean_clothing(text: str) -> str:
     return text.strip(" ,")
 
 
+def _body_phrase(body: Any, figure_note: Any = "") -> str:
+    body = _clean_text(body)
+    figure_note = _clean_text(figure_note)
+    if not body:
+        return figure_note
+    if not figure_note:
+        return f"{body} figure"
+    if "figure" in figure_note.lower():
+        return f"{body} build and {figure_note}"
+    return f"{body} figure with {figure_note}"
+
+
 def _single_caption_front(row: dict[str, Any]) -> dict[str, str]:
     caption = _clean_text(row.get("caption"))
     if not caption:
@@ -192,7 +204,7 @@ def _single_caption_front(row: dict[str, Any]) -> dict[str, str]:
     if not body_phrase:
         body = _clean_text(row.get("body_type") or row.get("body"))
         figure = _clean_text(row.get("figure"))
-        body_phrase = f"{body} figure with {figure}" if body and figure else f"{body} figure".strip()
+        body_phrase = _body_phrase(body, figure)
     front = f"{subject}, {age}, {body_phrase}, "
     if subject in ("woman", "man") and age and body_phrase and caption.startswith(front):
         try:
@@ -290,7 +302,7 @@ def _single_from_row(row: dict[str, Any], detail_level: str, keep_style: bool) -
     if not body_phrase:
         body = _clean_text(row.get("body_type") or row.get("body") or "")
         figure = _clean_text(row.get("figure"))
-        body_phrase = f"{body} figure with {figure}" if body and figure else f"{body} figure".strip()
+        body_phrase = _body_phrase(body, figure)
 
     skin = _row_value(row, "skin") or caption_front.get("caption_skin", "")
     hair = _row_value(row, "hair") or caption_front.get("caption_hair", "")
@@ -454,8 +466,48 @@ def _group_or_layout_from_row(row: dict[str, Any], detail_level: str, keep_style
     return _join_sentences(parts), "metadata(group_layout)"
 
 
+def _insta_of_pair_from_row(row: dict[str, Any], detail_level: str, keep_style: bool) -> tuple[str, str] | None:
+    if _clean_text(row.get("mode")).lower() != "insta/of":
+        return None
+    soft_row = row.get("softcore_row")
+    hard_row = row.get("hardcore_row")
+    if not isinstance(soft_row, dict) or not isinstance(hard_row, dict):
+        return None
+
+    hard_row_for_text = dict(hard_row)
+    options = row.get("options")
+    if isinstance(options, dict) and options.get("continuity") == "same_creator_same_room":
+        if soft_row.get("scene_text"):
+            hard_row_for_text["scene_text"] = soft_row["scene_text"]
+        if soft_row.get("composition"):
+            hard_row_for_text["composition"] = soft_row["composition"]
+
+    soft_text, _soft_method = _metadata_to_prose(soft_row, detail_level, keep_style)
+    hard_text, _hard_method = _metadata_to_prose(hard_row_for_text, detail_level, keep_style)
+    descriptor = _clean_text(row.get("shared_descriptor"))
+    cast_descriptors = row.get("shared_cast_descriptors")
+    if isinstance(cast_descriptors, list):
+        cast_descriptor_text = _human_join([_clean_text(item) for item in cast_descriptors if _clean_text(item)])
+    else:
+        cast_descriptor_text = _clean_text(cast_descriptors)
+
+    parts = []
+    if cast_descriptor_text:
+        parts.append(f"The shared cast descriptors are {cast_descriptor_text}")
+    elif descriptor:
+        parts.append(f"The shared creator descriptor is {descriptor}")
+    if soft_text:
+        parts.append(f"Softcore version: {soft_text}")
+    if hard_text:
+        parts.append(f"Hardcore version: {hard_text}")
+    if not parts:
+        return None
+    return _join_sentences(parts), "metadata(insta_of_pair)"
+
+
 def _metadata_to_prose(row: dict[str, Any], detail_level: str, keep_style: bool) -> tuple[str, str]:
     for builder in (
+        _insta_of_pair_from_row,
         _configured_cast_from_row,
         _single_from_row,
         _couple_from_row,

generate_prompt_batches.py

@@ -2823,6 +2823,18 @@ def figure_pool(mode: str) -> list:
     return FIGURE_CURVY  # 'curvy' default: voluptuous-leaning mix
 
 
+def make_body_phrase(body: str, figure_note: str = "") -> str:
+    body = str(body or "").strip()
+    figure_note = str(figure_note or "").strip()
+    if not body:
+        return figure_note
+    if not figure_note:
+        return f"{body} figure"
+    if "figure" in figure_note.lower():
+        return f"{body} build and {figure_note}"
+    return f"{body} figure with {figure_note}"
+
+
 def choose_woman(rng: random.Random, ethnicity: str = "any", no_plus: bool = False, no_black: bool = False):
     young = by_ethnicity(YOUNG_WOMEN, ethnicity)
     mature = by_ethnicity(MATURE_WOMEN, ethnicity)
@@ -2889,7 +2901,7 @@ def make_single(index: int, batch: int, rng: random.Random, gender: str, expr_de
         subject, age, body, skin, hair, eyes = choose(rng, men_pool)
         clothes = choose(rng, MEN_CLOTHES_MINIMAL if minimal else MEN_CLOTHES)
         figure_note = ""
-    body_phrase = f"{body} figure with {figure_note}" if figure_note else f"{body} figure"
+    body_phrase = make_body_phrase(body, figure_note)
 
     scene_slug, scene = choose(rng, SCENES)
     if poses == "evocative":

krea_formatter.py

@@ -118,6 +118,18 @@ def _row_value(row: dict[str, Any], key: str, labels: tuple[str, ...] = ()) -> s
     return ""
 
 
+def _body_phrase(body: Any, figure_note: Any = "") -> str:
+    body = _clean(body)
+    figure_note = _clean(figure_note)
+    if not body:
+        return figure_note
+    if not figure_note:
+        return f"{body} figure"
+    if "figure" in figure_note.lower():
+        return f"{body} build and {figure_note}"
+    return f"{body} figure with {figure_note}"
+
+
 def _single_caption_front(row: dict[str, Any]) -> dict[str, str]:
     caption = _strip_trigger(_clean(row.get("caption")), False)
     if not caption:
@@ -128,7 +140,7 @@ def _single_caption_front(row: dict[str, Any]) -> dict[str, str]:
     if not body:
         body_type = _clean(row.get("body_type") or row.get("body"))
         figure = _clean(row.get("figure"))
-        body = f"{body_type} figure with {figure}" if body_type and figure else f"{body_type} figure".strip()
+        body = _body_phrase(body_type, figure)
     front = f"{subject}, {age}, {body}, "
     if subject in ("woman", "man") and age and body and caption.startswith(front):
         try:
@@ -260,6 +272,11 @@ def _normal_row_to_krea(row: dict[str, Any], detail_level: str, style_mode: str)
 
 def _insta_pair_to_krea(row: dict[str, Any], detail_level: str, style_mode: str) -> tuple[str, str, str, str]:
     descriptor = _clean(row.get("shared_descriptor"))
+    cast_descriptors = row.get("shared_cast_descriptors")
+    if isinstance(cast_descriptors, list):
+        cast_descriptor_text = "; ".join(_clean(item) for item in cast_descriptors if _clean(item))
+    else:
+        cast_descriptor_text = _clean(cast_descriptors)
     soft = row.get("softcore_row") if isinstance(row.get("softcore_row"), dict) else {}
     hard = row.get("hardcore_row") if isinstance(row.get("hardcore_row"), dict) else {}
     soft_camera = _clean(row.get("softcore_camera_directive")) or _camera_phrase(soft)
@@ -274,9 +291,18 @@ def _insta_pair_to_krea(row: dict[str, Any], detail_level: str, style_mode: str)
     hard_cast_text = _clean(hard.get("cast_summary")) or (
         f"{hard_cast} adult women and {hard_men} adult men" if hard_cast or hard_men else ""
     )
+    same_room = options.get("continuity") == "same_creator_same_room"
+    hard_scene = soft.get("scene_text") if same_room and soft.get("scene_text") else hard.get("scene_text")
+    hard_composition = soft.get("composition") if same_room and soft.get("composition") else hard.get("composition")
+    soft_cast_descriptor_text = (
+        cast_descriptor_text
+        if options.get("softcore_cast") == "same_as_hardcore"
+        else f"Woman A / primary creator: {descriptor}"
+    )
 
     soft_parts = [
         f"A visibly adult creator, {descriptor}",
+        f"Shared cast descriptors: {soft_cast_descriptor_text}" if soft_cast_descriptor_text else "",
         f"shown in a {soft_level or 'softcore'} Insta/OF creator image",
         f"wearing {soft.get('item')}" if soft.get("item") else "",
         f"{soft.get('pose')}" if soft.get("pose") else "",
@@ -288,12 +314,13 @@ def _insta_pair_to_krea(row: dict[str, Any], detail_level: str, style_mode: str)
     ]
     hard_parts = [
         f"The same visibly adult creator, {descriptor}, is the visually central woman in a consensual explicit adult {hard_level or 'hardcore'} scene",
+        f"Shared cast descriptors: {cast_descriptor_text}" if cast_descriptor_text else "",
         f"all participants are 21+ and visibly adult; the cast includes {hard_cast_text}" if hard_cast_text else "all participants are 21+ and visibly adult",
         _clean(hard.get("role_graph")),
         f"The sexual action is {hard.get('item')}" if hard.get("item") else "",
-        f"set in {row.get('hardcore_row', {}).get('scene_text') or hard.get('scene_text')}" if hard.get("scene_text") else "",
+        f"set in {hard_scene}" if hard_scene else "",
         f"with {hard.get('expression')}" if hard.get("expression") else "",
-        f"framed as {hard.get('composition')}" if hard.get("composition") else "",
+        f"framed as {hard_composition}" if hard_composition else "",
         hard_camera,
         hard_style if detail_level != "concise" else "",
     ]

prompt_builder.py

@@ -1079,6 +1079,18 @@ def _merged_field(category: dict[str, Any], subcategory: dict[str, Any], item: A
     return default
 
 
+def _body_phrase(body: Any, figure_note: Any = "") -> str:
+    body = str(body or "").strip()
+    figure_note = str(figure_note or "").strip()
+    if not body:
+        return figure_note
+    if not figure_note:
+        return f"{body} figure"
+    if "figure" in figure_note.lower():
+        return f"{body} build and {figure_note}"
+    return f"{body} figure with {figure_note}"
+
+
 def _appearance_for_subject(
     rng: random.Random,
     subject_type: str,
@@ -1116,7 +1128,7 @@ def _appearance_for_subject(
         "skin": skin,
         "hair": hair,
         "eyes": eyes,
-        "body_phrase": f"{body} figure with {figure_note}",
+        "body_phrase": _body_phrase(body, figure_note),
         "figure": figure_note,
     }
 
@@ -1187,7 +1199,7 @@ def _configured_cast_context(women_count: int, men_count: int) -> dict[str, str]
 
 def _lettered(prefix: str, count: int) -> list[str]:
     letters = "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
-    return [f"{prefix} {letters[index]}" for index in range(max(0, count))]
+    return [f"{prefix.capitalize()} {letters[index]}" for index in range(max(0, count))]
 
 
 def _pick_distinct(rng: random.Random, items: list[str], count: int) -> list[str]:
@@ -2017,6 +2029,46 @@ def _insta_of_descriptor(row: dict[str, Any]) -> str:
     return ", ".join(str(piece).strip() for piece in pieces if piece and str(piece).strip())
 
 
+def _insta_of_descriptor_from_context(context: dict[str, Any]) -> str:
+    age = str(context.get("age") or "").strip()
+    age = " ".join(age.split())
+    age = age.removesuffix(" adults").removesuffix(" adult").strip()
+    subject = str(context.get("subject") or context.get("subject_type") or "person").strip()
+    pieces = [
+        f"{age} adult {subject}".strip(),
+        context.get("body_phrase"),
+        context.get("skin"),
+        context.get("hair"),
+        context.get("eyes"),
+    ]
+    return ", ".join(str(piece).strip() for piece in pieces if piece and str(piece).strip())
+
+
+def _insta_of_cast_descriptors(
+    primary_descriptor: str,
+    seed_config: dict[str, int],
+    seed: int,
+    row_number: int,
+    ethnicity: str,
+    figure: str,
+    no_plus_women: bool,
+    no_black: bool,
+    women_count: int,
+    men_count: int,
+) -> list[str]:
+    descriptors = [f"Woman A / primary creator: {primary_descriptor}"]
+    rng = _axis_rng(seed_config, "person", seed, row_number + 997)
+    for index in range(max(0, women_count - 1)):
+        label = chr(ord("B") + index)
+        context = _appearance_for_subject(rng, "woman", ethnicity, figure, no_plus_women, no_black)
+        descriptors.append(f"Woman {label}: {_insta_of_descriptor_from_context(context)}")
+    for index in range(max(0, men_count)):
+        label = chr(ord("A") + index)
+        context = _appearance_for_subject(rng, "man", ethnicity, figure, no_plus_women, no_black)
+        descriptors.append(f"Man {label}: {_insta_of_descriptor_from_context(context)}")
+    return descriptors
+
+
 def _insta_of_cast_phrase(women_count: int, men_count: int) -> str:
     context = _configured_cast_context(women_count, men_count)
     return context["cast_summary"]
@@ -2097,6 +2149,24 @@ def build_insta_of_pair(
     )
 
     descriptor = _insta_of_descriptor(soft_row)
+    cast_descriptors = _insta_of_cast_descriptors(
+        descriptor,
+        parsed_seed_config,
+        seed,
+        row_number,
+        ethnicity,
+        figure,
+        no_plus_women,
+        no_black,
+        hard_women_count,
+        hard_men_count,
+    )
+    cast_descriptor_text = "; ".join(cast_descriptors)
+    soft_cast_descriptor_text = (
+        cast_descriptor_text
+        if options["softcore_cast"] == "same_as_hardcore"
+        else f"Woman A / primary creator: {descriptor}"
+    )
     platform_style = INSTA_OF_PLATFORM_STYLES[options["platform_style"]]
     soft_level = INSTA_OF_SOFT_LEVELS[options["softcore_level"]]
     hard_level = INSTA_OF_HARDCORE_LEVELS[options["hardcore_level"]]
@@ -2121,6 +2191,7 @@ def build_insta_of_pair(
     soft_prompt = (
         f"Insta/OF softcore mode: {platform_style}. Shared primary creator descriptor: {descriptor}. "
         f"Softcore setup: {soft_level}. Cast continuity: {soft_cast}. "
+        f"Shared cast descriptors: {soft_cast_descriptor_text}. "
         f"Outfit: {soft_row['item']}. Pose: {soft_row['pose']}. Setting: {soft_row['scene_text']}. "
         f"Facial expression: {soft_row['expression']}. Composition: {soft_row['composition']}. "
         f"{soft_camera_sentence}"
@@ -2130,6 +2201,7 @@ def build_insta_of_pair(
     hard_prompt = (
         f"Insta/OF hardcore mode: {platform_style}. Shared primary creator descriptor: {descriptor}. "
         f"Hardcore setup: {hard_level}. Cast: {hard_cast}. "
+        f"Shared cast descriptors: {cast_descriptor_text}. "
         "Apply the shared descriptor to the most visually central woman, keeping her continuous with the softcore version. "
         f"Role graph: {hard_row['role_graph']} Sexual scene: {hard_row['item']}. "
         f"Setting: {hard_scene}. Facial expressions: {hard_row['expression']}. Composition: {hard_composition}. "
@@ -2159,6 +2231,7 @@ def build_insta_of_pair(
         "mode": "Insta/OF",
         "options": options,
         "shared_descriptor": descriptor,
+        "shared_cast_descriptors": cast_descriptors,
         "softcore_prompt": soft_prompt,
         "hardcore_prompt": hard_prompt,
         "softcore_negative_prompt": soft_negative,