画面会恐怖或血腥吗？

不会。配方采用脚踏实地的民间传说氛围，用雾、萤火虫和提灯建立情绪，不要求血液、伤口或怪物脸。

什么照片最合适？

光线均匀的四分之三身或全身照片更合适，因为斗篷和帽子需要空间。避免强逆光把脸压黑。

全部万圣节造型

万圣节 AI 工作室

森林女巫提灯肖像

生成苔绿斗篷、歪檐帽、萤火虫、雨后森林和一盏暖灯的自然系女巫肖像。

生成这个造型把造型做成视频

森林女巫

适合喜欢氛围感而不是卡通造型的人。人物保持可识别，森林与提灯承担大部分节日情绪。

苔绿面料、亚麻与皮革层次、植物小挂件和歪檐帽替换原服装。湿根、萤火虫与提灯补足空间层次，同时不压过脸部。

适合

情绪化秋季肖像
森林系或 cottagecore 变装概念
薄雾与萤火虫的轻微动画

提示词本身就是产品

可以直接复制完整配方，也可以点按钮进入已填好提示词的编辑器。身份、服装、场景、光线和失败项分开描述，复验时一次只改一个变量。

图片提示词

Use the uploaded photo as the identity reference. Preserve the person's recognizable face, skin texture, facial proportions, age, pose, and natural anatomy. Create an original forest-witch Halloween portrait with a dark moss-green cloak, layered linen and leather, small botanical charms, and a wide crooked brim hat. Keep the styling grounded and handcrafted rather than cartoonish. Place the subject in an ancient foggy woodland just after rain, with wet roots, fireflies, and a single warm lantern. Use photorealistic editorial photography, soft cool ambient light, warm lantern fill, realistic skin and fabric. Keep the same number of people as the uploaded photo. Avoid altered identity, extra limbs, distorted hands, text, logos, and watermarks.

视频提示词

Use the provided portrait as the exact first frame. Create a five-second atmospheric Halloween forest-witch animation. Preserve identity, costume, hat, pose, background, and lighting. The subject looks slowly toward the lantern as fireflies move through the mist. Moist fabric and cloak edges shift in a quiet breeze, and the lantern light flickers gently. Use a minimal camera push-in, realistic physics, stable anatomy, and no cuts. Do not change the face or costume, add creatures, introduce text, or add watermarks.

提示词保留英文，因为图片和视频模型对英文指令的理解最稳定。进入编辑器后仍可自行修改。