Image Director
Use when the user wants to做一整套图片、组图、卡片图、轮播图、配图包或电商套图.
Install
npx skills add Jamailar/Beav --skill image-director -g -a claude-code -y- Repository
- Jamailar/Beav2.2k
- Category
- Media
- Updated
- 3d ago
- License
- See repository
- Context cost
- Mediumabout 3,000 tokens
- Runs scripts
- No, instructions only
What it does
Use when the user wants to做一整套图片、组图、卡片图、轮播图、配图包或电商套图. First identify the user's motive and final goal, then choose the right image-set type and ordering strategy, lock the batch-level subject anchor, keep the style guide concise, define text placement and image details for each card, and always show the full batch plan to the user first. Only after the user explicitly agrees may you call Operate(resource="image", operation="generate", input={ ... }) for batch generation. Never submit multi-image generation without user approval.
Image Director
Use this skill only for multi-image work.
If the user needs just one image, do not use this workflow.
When To Use
Use this skill when the request implies a coordinated image set, for example:
- 一次生成多张配图
- 小红书组图 / 轮播图 / 多卡封面
- 连续步骤图 / 场景序列图
- 同一主体、同一风格下的多张海报或插图
- 把文章做成卡片 / 把内容改成图文卡片
- 做图解卡片 / 知识卡片 / 图文卡片 / 演示卡片
- 做电商套图 / 商品套图 / 商品详情图 / 转化配图
中文触发词与入口词
看到下面这些中文表达时，应优先判断是否进入本技能，而不是把它当成单张图片任务：
电商套图/商品套图/商品配图/商品详情图/电商配图文章卡片/文章改卡片/把文章做成卡片/图文卡片图解卡片/知识卡片/知识图解/拆解卡片演示卡片/小红书演示卡片/演示图文/轮播演示卡组图/多张图/多卡/轮播图/成套配图
这些入口词的默认归类如下：
电商套图->电商配图文章卡片-> 优先归到知识卡片套图或小红书文字卡片，根据文章是否以观点表达为主来判断图解卡片-> 优先归到知识卡片套图；如果明显要求角色化、分镜化解释，可归到知识图解漫画演示卡片-> 归到小红书文字卡片或演示型轮播卡片，重点是步骤展示、卖点演示、前后对比或操作过程
Planning Priorities
When planning a batch, follow this priority order:
- Determine the user's motive and the set's business/content goal first.
- Lock subject consistency second.
- Lock the shared style direction third.
- Spend the most detail on visible text, layout positions, and image-level visual details.
This means:
- 先决定这套图是为了吸引点击、解释知识、展示产品，还是促进转化
- 主体一致性高于“每张都很花哨”
- 套图统一感高于单张炫技
- 细节说明高于冗长风格辞藻
Do not treat style language as the main payload. Modern image models usually already produce decent aesthetics from a short style anchor. Use the extra detail budget on identity consistency, composition constraints, text content, text position, product placement, props, background details, and must-keep visual cues.
Core Mission
The core mission of this skill is:
- Determine the exact content details of every image.
- Ensure the whole set stays visually unified and never feels fragmented.
- Produce a clear batch plan for user review before any multi-image generation call happens.
- Wait for explicit user approval before calling
image.generate.
