feat(baoyu-article-illustrator): add batch generation policy with configurable batch size

skills/baoyu-article-illustrator/references/config/first-time-setup.md

@@ -129,12 +129,15 @@ preferred_style:
 default_output_dir: imgs-subdir  # same-dir | imgs-subdir | illustrations-subdir | independent
 language: null
 preferred_image_backend: auto
+generation_batch_size: 4
 custom_styles: []
 ---
 ```
 
 `preferred_image_backend: auto` is the baked-in default — first-time setup does not ask about it. The `## Image Generation Tools` rule in SKILL.md then picks the runtime-native tool (Codex `imagegen`, Hermes `image_generate`, etc.) when available, and falls back to installed backends.
 
+`generation_batch_size: 4` is the baked-in default for batch rendering. The current user request may override it for one run.
+
 ## Modifying Preferences Later
 
 See the `## Changing Preferences` section in `SKILL.md` for the canonical list of common edits (pin backend, change defaults, retrigger setup). Full schema: `preferences-schema.md`.

skills/baoyu-article-illustrator/references/config/preferences-schema.md

@@ -28,6 +28,8 @@ default_output_dir: null  # same-dir|illustrations-subdir|independent
 
 preferred_image_backend: auto  # auto|ask|<backend-id>
 
+generation_batch_size: 4       # 1-8, used when backend/runtime supports batch or parallel generation
+
 custom_styles:
   - name: my-style
     description: "Style description"
@@ -55,6 +57,7 @@ custom_styles:
 | `language` | string | null | Output language (null = auto-detect) |
 | `default_output_dir` | enum | null | Output directory preference (null = ask each time) |
 | `preferred_image_backend` | string | `auto` | Image backend selection. `auto` = prefer runtime-native tool, fall back to the only installed backend, ask if multiple non-native are present. `ask` = always confirm on every run. `<backend-id>` (e.g., `codex-imagegen`, `baoyu-imagine`, `image_generate`) = pin this backend when available; fall back to `auto` when it isn't. Absent = `auto`. Resolution logic is documented in `SKILL.md`'s `## Image Generation Tools` section. |
+| `generation_batch_size` | int | 4 | Number of images to dispatch per batch when the backend has native batch support or the runtime can issue parallel generation calls. Clamp invalid values to 1-8. Current user request overrides this value. |
 | `custom_styles` | array | [] | User-defined styles |
 
 ## Position Options
@@ -118,6 +121,8 @@ language: zh
 
 preferred_image_backend: codex-imagegen
 
+generation_batch_size: 4
+
 custom_styles:
   - name: corporate
     description: "Professional B2B style"

skills/baoyu-article-illustrator/references/usage.md

@@ -18,6 +18,9 @@
 # Specify density
 /baoyu-article-illustrator path/to/article.md --density rich
 
+# Generate up to 4 images in parallel after prompts are saved
+/baoyu-article-illustrator path/to/article.md --batch-size 4
+
 # Direct content input (paste mode)
 /baoyu-article-illustrator
 [paste content]
@@ -31,6 +34,7 @@
 | `--style <name>` | Visual style (see references/styles.md) |
 | `--preset <name>` | Shorthand for type + style combo (see [references/style-presets.md](references/style-presets.md)) |
 | `--density <level>` | Image count: minimal / balanced / rich |
+| `--batch-size <n>` | Temporary generation batch size for this run. Default: `generation_batch_size` from EXTEND.md, otherwise 4. Clamp to 1-8. |
 
 ## Input Modes
 

skills/baoyu-article-illustrator/references/workflow.md

@@ -335,8 +335,11 @@ Prompt Files:
 **DO NOT** pass ad-hoc inline text to `--prompt` without first saving prompt files. The generation command should either use `--promptfiles prompts/NN-{type}-{slug}.md` or read the saved file content for `--prompt`.
 
 **Execution choice**:
-- If multiple illustrations already have saved prompt files and the task is now plain generation, prefer the chosen backend's batch interface (if it offers one); otherwise generate sequentially
-- Use subagents only when each illustration still needs separate prompt rewriting, style exploration, or other per-image reasoning before generation
+- If multiple illustrations already have saved prompt files and the task is now plain generation, use batch generation by default.
+- Prefer the chosen backend's native batch / multi-task interface when available.
+- If the backend has no native batch interface but the runtime can issue parallel tool calls, dispatch up to `generation_batch_size` tasks at a time. Default: `4`. The current user request overrides EXTEND.md.
+- Generate sequentially only when neither backend batch nor runtime parallel calls are available.
+- Use subagents only when each illustration still needs separate prompt rewriting, style exploration, or other per-image reasoning before generation. Do not use subagents just to parallelize rendering.
 
 **CRITICAL - References in Frontmatter**:
 - Only add `references` field if files ACTUALLY EXIST in `references/` directory
@@ -394,12 +397,18 @@ Add: `Include a subtle watermark "[content]" at [position].`
 
 ### 5.5 Generate
 
-1. For each illustration:
-   - **Backup rule**: If image file exists, rename to `NN-{type}-{slug}-backup-YYYYMMDD-HHMMSS.md`
-   - If references with `direct` usage: include `--ref` parameter
-   - Generate image
-2. After each: "Generated X/N"
-3. On failure: retry once, then log and continue
+1. Build a generation task list from saved prompt files:
+   - `prompt_file`: `{output-dir}/prompts/NN-{type}-{slug}.md`
+   - `output_file`: `{output-dir}/NN-{type}-{slug}.png`
+   - `aspect_ratio`: from prompt frontmatter or prompt body
+   - `refs`: only verified `direct` references from prompt frontmatter
+2. **Backup rule**: Before dispatching a task, if its output image already exists, rename it to `NN-{type}-{slug}-backup-YYYYMMDD-HHMMSS.{ext}`.
+3. Dispatch tasks in batches:
+   - Native batch backend: send all eligible tasks, or chunks of `generation_batch_size` if the backend has a practical limit.
+   - Runtime parallel calls: issue up to `generation_batch_size` image calls concurrently, then continue with the next chunk.
+   - Sequential fallback: process one task at a time.
+4. After each completed task, record: "Generated X/N: filename".
+5. On failure: retry the failed task once from the same saved prompt file. Keep successful outputs and continue.
 
 ---