chore: release v1.18.3

.claude-plugin/marketplace.json

@@ -6,7 +6,7 @@
   },
   "metadata": {
     "description": "Skills shared by Baoyu for improving daily work efficiency",
-    "version": "1.18.2"
+    "version": "1.18.3"
   },
   "plugins": [
     {

CHANGELOG.md

@@ -2,6 +2,14 @@
 
 English | [中文](./CHANGELOG.zh.md)
 
+## 1.18.3 - 2026-01-23
+
+### Documentation
+- `baoyu-comic`: improves character reference handling with explicit Strategy A/B selection—Strategy A uses `--ref` parameter for skills that support it, Strategy B embeds character descriptions in prompts for skills that don't. Includes concrete code examples for both approaches.
+
+### Fixes
+- `baoyu-image-gen`: removes unsupported Gemini models (`gemini-2.0-flash-exp-image-generation`, `gemini-2.5-flash-preview-native-audio-dialog`) from multimodal model list.
+
 ## 1.18.2 - 2026-01-23
 
 ### Refactor

CHANGELOG.zh.md

@@ -2,6 +2,14 @@
 
 [English](./CHANGELOG.md) | 中文
 
+## 1.18.3 - 2026-01-23
+
+### 文档
+- `baoyu-comic`：改进角色参考处理流程，新增明确的 Strategy A/B 选择逻辑——Strategy A 使用 `--ref` 参数（适用于支持该参数的技能），Strategy B 将角色描述嵌入提示词（适用于不支持的技能）。包含两种方法的具体代码示例。
+
+### 修复
+- `baoyu-image-gen`：从多模态模型列表中移除不支持的 Gemini 模型（`gemini-2.0-flash-exp-image-generation`、`gemini-2.5-flash-preview-native-audio-dialog`）。
+
 ## 1.18.2 - 2026-01-23
 
 ### 重构

skills/baoyu-comic/SKILL.md

@@ -380,9 +380,50 @@ With confirmed storyboard + art style + tone + aspect ratio:
 
 **5.2 Generate Comic Pages**:
 
+**CRITICAL: Character Reference is MANDATORY** for visual consistency across all pages.
+
+**Before generating any page**:
+1. Read the image generation skill's SKILL.md
+2. Check if it supports reference image input (`--ref`, `--reference`, etc.)
+3. Choose the appropriate strategy below
+
+**Character Reference Strategy**:
+
+| Skill Capability | Strategy | Action |
+|------------------|----------|--------|
+| Supports `--ref` | **Strategy A** | Pass `characters/characters.png` with EVERY page |
+| Does NOT support `--ref` | **Strategy B** | Prepend character descriptions to EVERY prompt |
+
+**Strategy A: Using `--ref` parameter** (e.g., baoyu-image-gen)
+
+```bash
+# Each page generation MUST include --ref
+npx -y bun ${SKILL_DIR}/../baoyu-image-gen/scripts/main.ts \
+  --promptfiles prompts/01-page-xxx.md \
+  --image 01-page-xxx.png \
+  --ar 3:4 \
+  --ref characters/characters.png
+```
+
+**Strategy B: Embedding character descriptions in prompt**
+
+When skill does NOT support reference images, create combined prompt files:
+
+```markdown
+# prompts/01-page-xxx.md (with embedded character reference)
+
+## Character Reference (maintain consistency)
+[Copy relevant sections from characters/characters.md here]
+- 大雄: Japanese boy, round glasses, yellow shirt, navy shorts...
+- 哆啦A梦: Round blue robot cat, white belly, red nose, golden bell...
+
+## Page Content
+[Original page prompt here]
+```
+
 **For each page (cover + pages)**:
-1. Save prompt to `prompts/NN-{cover|page}-[slug].md` (in user's preferred language)
+1. Save prompt to `prompts/NN-{cover|page}-[slug].md`
-2. Generate image using confirmed art style and tone guidelines
+2. Generate image using Strategy A or B (based on skill capability)
 3. Report progress after each generation
 
 **Watermark Application** (if enabled in preferences):
@@ -395,22 +436,6 @@ Ensure watermark does not overlap speech bubbles or key action.
 ```
 Reference: `references/config/watermark-guide.md`
 
-**Image Generation Skill Selection**:
-- Check available image generation skills
-- If multiple skills available, ask user preference
-
-**Character Reference Handling** (IMPORTANT for visual consistency):
-
-1. Read the selected image generation skill's SKILL.md
-2. Check if it supports reference image input (look for parameters like `--ref`, `--reference`, `--image-ref`, etc.)
-
-| Skill Support | Action |
-|---------------|--------|
-| **Supports reference image** | Pass `characters/characters.png` using the skill's reference parameter |
-| **Does NOT support reference image** | Prepend `characters/characters.md` content to each page prompt |
-
-Always verify the exact parameter name from the skill's documentation before calling.
-
 **Session Management**:
 If image generation skill supports `--sessionId`:
 1. Generate unique session ID: `comic-{topic-slug}-{timestamp}`

skills/baoyu-image-gen/scripts/main.ts

@@ -24,8 +24,6 @@ type CliArgs = {
 
 const GOOGLE_MULTIMODAL_MODELS = [
   "gemini-3-pro-image-preview",
-  "gemini-2.0-flash-exp-image-generation",
-  "gemini-2.5-flash-preview-native-audio-dialog",
 ];
 
 const GOOGLE_IMAGEN_MODELS = ["imagen-3.0-generate-002", "imagen-3.0-generate-001"];