← 返回 Skills 市场
zoucdr

拟人动物化妆变身视频

作者 parallel world · GitHub ↗ · v1.0.1 · MIT-0
cross-platform ⚠ suspicious
152
总下载
0
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install makeup-transform-video
功能描述
生成拟人动物化妆变身短视频。素颜到惊艳的快节奏卡点反差内容,支持日常妆、约会妆、舞台妆、角色仿妆等场景,一句话描述直接出片。
使用说明 (SKILL.md)

化妆变身视频生成

素颜到惊艳的反差视频——核心是反差感和卡点节奏,适合变美赛道、创意拟人内容、节日妆容展示。拟人动物角色(小狐狸/猫咪/兔兔)完成从素颜到精致的完整变身过程,5 秒内完成情绪弧线。

依赖:本目录 scripts/video_gen.js + 运行环境需设置 WERYAI_API_KEY + Node.js 18+。不依赖其他 Cursor 技能。


默认参数

参数
模型 KLING_V3_0_PRO
比例 9:16(固定)
时长 5 秒(duration: 5,快节奏卡点)
音频 关闭
视觉风格 正面近景/特写,柔和美妆布光,妆前自然光偏冷,妆后暖光增强高光,快节奏切换

API 合法性(默认 KLING_V3_0_PRO:文生 duration5 / 10 / 15aspect_ratio9:16、1:1、16:9;图生 aspect_ratio9:16、16:9、1:1resolution 字段,请求勿传。快档若用 VEO:文生 VEO_3_1_FAST、图生 CHATBOT_VEO_3_1_FAST,且 duration 固定 8aspect_ratio9:1616:9。完整全模型表见 weryai-model-capabilities.md


素颜到精致变身

使用前准备: 确认以下三项即可生成:

  • 角色类型(小狐狸 / 猫咪 / 兔兔 / 小熊 / 可自定义任何动物)
  • 妆容风格(日常妆 / 约会妆 / 舞台妆 / 圣诞节日妆 / 角色仿妆 / 夸张舞台妆)
  • 反差强度(自然提升感 / 巨大反差 / 夸张角色变身)

生成时构建包含「素颜→化妆过程→完妆特写」完整弧线的 prompt,强调高光、眼妆细节和完妆后的气质变化。提交前展示参数确认:

必须以表格展示所有参数,等待用户明确确认后再提交生成:

📋 即将生成,请确认以下参数:

参数 本次值 说明
model KLING_V3_0_PRO 优档默认;快档:文生 VEO_3_1_FAST、图生 CHATBOT_VEO_3_1_FASTduration 固定 8);均衡→KLING_V3_0_STA;可直接指定模型名
aspect_ratio 9:16 默认 KLING:仅 9:16、1:1、16:9;换模见 ../references/weryai-model-capabilities.md
duration 5s KLING 系:5 / 10 / 15;切 VEO 快档:duration 仅 8
generate_audio false 是否自动生成音频
prompt (本次构建的描述摘要) 如需调整请告知
循环衔接 回复"循环"可开启 seamless loop

直接回复**"确认"**开始生成,或告知需要修改的项目。

示例 Prompt:

A small elegant fox with amber eyes starts with a clean bare face under soft natural window light, medium close-up frontal shot, the fox applies foundation with a damp sponge blending outward, then precise liquid liner wing tips extend dramatically, rich burgundy eyeshadow blended in the crease, final glossy red lip applied in one smooth stroke, the camera slowly pushes in as the fox turns to face forward — the transformation from plain to breathtaking dinner-party queen, warm beauty ring light catches highlighter on cheekbones, fast rhythm cut at final reveal, 9:16 portrait orientation

A white cat with blue eyes applies full Christmas holiday makeup, close-up medium shot, bare-faced start with cool neutral lighting transitions to warm sparkle as glitter is pressed onto the inner corners, red and green eyeshadow gradient applied symmetrically, tiny gold star gems placed at the outer corners, deep berry lips finished with gloss, final reveal: the cat turns forward with a slow smile, sparkles catch the warm LED ring light, festive atmosphere, soft bokeh background with blurred fairy lights

A gray bunny transforms with an exaggerated theatrical stage look in 3 compressed minutes, starting from entirely bare, each step dramatically cuts forward: base concealing all natural features, graphic black liner extending beyond the natural eye shape, neon orange shadow blended to the brow, rhinestones scattered across the lid, the final frame catches the bunny in full drag-inspired stage makeup under pure white theater lighting, jaw-dropping before-after contrast in a single vertical frame

预期效果: 妆容变化过程清晰,眼影层次和高光质感在视频中可见,最终完妆帧的气质与素颜帧形成强烈反差,适合变美/创意账号的卡点剪辑素材。


分步精致妆容过程

专注于单个妆步的极致细节,适合化妆教程向内容——强调产品质地、上妆手势、每一步的即时效果变化,而非整体变身弧线。

使用前准备: 明确想展示的妆步类型(底妆服帖感 / 眼线精准收尾 / 口红上色瞬间 / 睫毛膏刷层叠加 / 修容阴影塑形)以及角色造型偏好。

生成流程:收集妆步类型和细节偏好 → 构建强调手势动作和产品质地的 prompt → 确认参数 → 执行 node {baseDir}/scripts/video_gen.js wait --json '…'(字段与确认表一致)

生成前会展示参数,等待确认。

示例 Prompt:

Extreme macro close-up of a small fox applying a single swipe of deep cherry-red lipstick across full lips, the bullet tip glides across the lower lip first then upper, leaving a perfect saturated pigment trail, the color payoff is immediate and glossy under ring light, the fox presses lips together once — lipstick perfectly transfers both sides, warm backlight catches the gloss sheen, static locked-off shot with shallow depth of field blurring background, slow motion 120fps, ASMR application sound

参数配置:

参数
model KLING_V3_0_PRO
aspect_ratio 9:16
duration 5
generate_audio false

注意事项

妆容层次感:在 prompt 里明确列出 2-3 个妆步,如 applies foundation → adds eyeshadow → finishes with lip 的顺序描述,比笼统写"化妆"生成的妆容层次感更丰富。

反差强化:在 prompt 中用 bare-faced start + jaw-dropping transformation + before-after contrast 三段式结构,确保模型理解这是一个有戏剧性弧线的视频,而不是静态美妆展示。

光线切换:妆前用 cool natural window light,妆后用 warm ring light catches highlighter,光线温度的转变会让反差感在视觉上更强烈。

安全使用建议
This skill appears to be what it says: a Node CLI that calls a WeryAI video generation API. Before installing or running it, do the following: (1) Confirm you are willing to provide a WERYAI_API_KEY (the skill needs this though the registry metadata omitted it). Use a key tied to an account you control and understand usage/charges. (2) Do not set WERYAI_BASE_URL or WERYAI_MODELS_BASE_URL to unknown domains — those env vars can redirect requests to other servers. (3) Review example prompts to avoid embedding personal/private data or copyrighted imagery in prompts or upload images you don't own. (4) Run a dry-run (the script supports --dry-run) first to observe behavior. The main discrepancy is the missing required env var entry in registry metadata — that is likely an oversight but worth confirming with the skill author if you need stricter guarantees.
功能分析
Type: OpenClaw Skill Name: makeup-transform-video Version: 1.0.1 The skill bundle is a legitimate tool for generating makeup transformation videos using the WeryAI API. The Node.js script (scripts/video_gen.js) acts as a standard CLI client that handles task submission and status polling via the WeryAI endpoints (api.weryai.com). The instructions in SKILL.md are well-aligned with the stated purpose, providing clear guidance for the AI agent to construct prompts and confirm parameters with the user before execution, with no evidence of malicious intent or data exfiltration.
能力评估
Purpose & Capability
Name/description (makeup transform videos) match the provided code and runtime instructions: the script submits text/image-to-video generation requests to a WeryAI endpoint and polls for results. Requiring Node and a WeryAI API key is coherent with this purpose.
Instruction Scope
SKILL.md instructs the agent to collect user parameters, display them for confirmation, then run the included CLI (node scripts/video_gen.js ...) to submit generation jobs. The instructions do not ask the agent to read unrelated files or access unrelated system credentials.
Install Mechanism
No install spec is present (instruction + a small CLI script). There is no network-based installer, no archive extraction, and the script has no npm dependencies. Risk from install mechanism is low.
Credentials
The SKILL.md and script require WERYAI_API_KEY (and accept optional WERYAI_BASE_URL, WERYAI_MODELS_BASE_URL, and poll-timing env vars), but the registry metadata lists no required env vars. The missing declaration is a discrepancy — the API key is necessary and should be explicitly declared. The optional BASE_URL overrides could be misused if set to an untrusted endpoint, so users should not set them to unknown domains.
Persistence & Privilege
The skill is user-invocable, not always-included, and does not request elevated agent privileges or attempt to modify other skills. Autonomous invocation is allowed (platform default) but not combined with any additional privileged behavior.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install makeup-transform-video
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /makeup-transform-video 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.1
No changes detected in this version (1.0.1) compared to 1.0.0. - No file changes found. - SKILL.md and all other files remain unchanged.
v1.0.0
Initial release of makeup-transform-video. - Generate short, high-impact makeup transformation videos for anthropomorphic animal characters. - Supports various makeup styles (everyday, date, stage, cosplay, festive) and custom animal types. - Workflow ensures user confirms all parameters (model, aspect ratio, duration, etc.) before video is generated, using a clear confirmation table. - Prompts and parameters optimized for dramatic before/after contrast, rhythm, and clarity of makeup steps. - Designed for fast-paced, short-form video (default 5 seconds, 9:16 vertical, no audio).
元数据
Slug makeup-transform-video
版本 1.0.1
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 2
常见问题

拟人动物化妆变身视频 是什么?

生成拟人动物化妆变身短视频。素颜到惊艳的快节奏卡点反差内容,支持日常妆、约会妆、舞台妆、角色仿妆等场景,一句话描述直接出片。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 152 次。

如何安装 拟人动物化妆变身视频?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install makeup-transform-video」即可一键安装,无需额外配置。

拟人动物化妆变身视频 是免费的吗?

是的,拟人动物化妆变身视频 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

拟人动物化妆变身视频 支持哪些平台?

拟人动物化妆变身视频 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 拟人动物化妆变身视频?

由 parallel world(@zoucdr)开发并维护,当前版本 v1.0.1。

💬 留言讨论