From short-video-director
Batch generates reference images for story episode assets. Reads config.md for image model, collects assets from outline.md, skips existing images, delegates to creator-image-{model} skill.
npx claudepluginhub wddxh/shortvideodirectorThis skill is limited to using the following tools:
- `config.md` — 必须读取（获取图像模型值）
Generates reference images for assets using Dreamina CLI: extracts prompts from MD files, handles login checks, classifies basic/variant assets, polls pending tasks, logs timeouts to pending.json.
Generates images from structured JSON prompts via Python script execution. Supports reference images and aspect ratios for characters, scenes, products, visuals.
Generates and iterates on AI images with consistent style and subject identity using image-sprout CLI projects, references, guides, and OpenRouter integration.
Share bugs, ideas, or general feedback.
config.md — 必须读取（获取图像模型值）story/episodes/$ARGUMENTS[0]/outline.md — 必须读取（从「本集资产清单」获取资产列表）$ARGUMENTS[0] — 当前集数（如 ep01）读取 config 中的图像模型配置，收集需要生成图片的资产列表，委托给对应的模型 skill 执行图片生成。
config.md，获取 图像模型 值（使用 Bash 调用 bash scripts/read-config.sh "图像模型" 获取图像模型值）none → 输出"图像模型未配置，跳过图片生成"并结束story/episodes/{集数}/outline.md 中的 ## 本集资产清单，收集所有资产文件路径（包括新增资产和已有资产）bash scripts/asset-to-image-path.sh "{资产路径}" 转换路径，然后检查文件是否存在），已存在则跳过creator-image-{图像模型值} skill，传递参数：需要生成图片的资产路径列表（空格分隔，每个路径用引号包裹）