smart_image_gen v0.7.7: enforce style inheritance for edit_image

Vision-capable LLMs misclassify rendered subjects when picking a
style — observed: model called juggernaut for an edit on a furry-il
generation because the rendered character looked 'photoreal-ish' to
its vision encoder. Each visual judgment is independent so styles
flip mid-chat.

Flipped resolution order in edit_image so inheritance from the prior
generate_image / edit_image call DOMINATES the LLM's explicit style
arg. The LLM's choice only wins when there's nothing to inherit
(first edit in a chat, fresh user upload). Workaround for legitimate
style changes is starting a new chat.

System prompt updated to match: tells the LLM that style inheritance
is enforced, that passing style on follow-up calls is ignored, and
that user requests for style change require a new chat.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

deployments/ai-stack/openwebui-tools/smart_image_gen.py

@@ -1,7 +1,7 @@
 """
 title: Smart Image Generator & Editor (ComfyUI)
 author: ai-stack
-version: 0.7.6
+version: 0.7.7
 description: Generate or edit images via ComfyUI with automatic SDXL
     checkpoint routing. Two methods — generate_image (txt2img) and
     edit_image (img2img on the user's most recently attached image). The
@@ -872,12 +872,15 @@ class Tools:
 
         Pick `style` for the DESIRED OUTPUT, not the input image.
 
-        Style resolution order: explicit `style` arg → inherited from the
-        most recent prior generate_image / edit_image call in this
-        conversation → keyword detection on `prompt`. Omit `style` to
-        let the tool inherit from the previous call automatically — it
-        usually picks the right thing for follow-up edits on an image
-        the LLM just generated.
+        Style resolution order: inherited from the most recent prior
+        generate_image / edit_image call in this conversation (DOMINANT)
+        → explicit `style` arg → keyword detection on `prompt`.
+        Inheritance dominates because vision LLMs misclassify subjects
+        in the rendered output (e.g. picking 'juggernaut' on a
+        'furry-il' source). For follow-up edits on an image you
+        generated earlier, omit `style` entirely — the tool reuses the
+        established style automatically. The user can start a new chat
+        if they want a different style.
 
         :param prompt: What the changed area should look like.
             Tool auto-prepends quality tags — don't include those.
@@ -892,12 +895,13 @@ class Tools:
         :param seed: 0 to randomize, otherwise specific.
         :return: Markdown image of the result, or an error if no image is attached.
         """
-        # Resolve style with explicit > inherited-from-prior-call > keyword.
-        # Inheritance covers the common case where the LLM is editing an
-        # image it already generated and forgets to set style — without it,
-        # neutral edit prompts ("bigger", "glowing eyes") fall through to
-        # the keyword router and get the wrong checkpoint.
-        chosen = style or _inherited_style(__messages__) or _route_style(prompt)
+        # Resolve style — inheritance DOMINATES for edits. Vision LLMs
+        # misclassify subject types (observed in the wild: juggernaut
+        # picked for a furry-il source because the model thought the
+        # rendered character looked "photoreal-ish"). When there's a
+        # prior tool call in this chat, use the same style; the user's
+        # workaround for genuine style changes is a fresh chat.
+        chosen = _inherited_style(__messages__) or style or _route_style(prompt)
         settings = STYLES.get(chosen)
         if not settings:
             return f"Unknown style '{chosen}'. Available: {', '.join(STYLES.keys())}"