image-face-swap
Skill33 starsChanged 5 months agoMITView on GitHub
双图换脸/换头合成,将源图人脸替换到目标场景图上,保留目标图构图、身体和背景并自动做肤色融合与光影匹配。仅在用户明确提供两张图,并表达换脸、换头、把 A 的脸换到 B 上、face swap 或明确的 P 脸合成意图时触发。执行时会使用本地 Meitu OpenAPI 凭证授权,并将源人脸图和目标场景图发送到 Meitu 外部服务处理。
- Reads credentials
- Installs packages
In this repository55 other files it gives its agents.
- meitu-toolsSkill
- SKILL.mdSkill
- audio-music-generateSkill
- audio-song-generateSkill
- image-background-replaceSkill
- image-cutoutSkill
- image-denoise-enhanceSkill
- image-editSkill
- image-element-removeSkill
- image-grid-splitSkill
- image-id-photo-generateSkill
- image-lowlight-enhanceSkill
- image-outfit-swapSkill
- image-portrait-generateSkill
- image-poster-generateSkill
- image-searchSkill
- image-style-transferSkill
- image-superres-enhanceSkill
- image-text-replaceSkill
- image-to-videoSkill
- image-transformSkill
- meitu-beautySkill
- meitu-carouselSkill
- meitu-cutoutSkill
- meitu-game-2d-assetsSkill
- meitu-id-photoSkill
- meitu-image-adaptSkill
- meitu-image-fixSkill
- meitu-posterSkill
- meitu-product-swapSkill
- meitu-product-viewSkill
- meitu-stickersSkill
- meitu-upscaleSkill
- meitu-video-danceSkill
- meitu-visual-meSkill
- text-code-editSkill
- text-codeSkill
- text-generateSkill
- text-to-imageSkill
- text-to-videoSkill
- video-audio-addSkill
- video-canvas-expandSkill
- video-content-replaceSkill
- video-denoise-enhanceSkill
- video-effect-applySkill
- video-element-removeSkill
- video-framerate-enhanceSkill
- video-logo-addSkill
- video-lowlight-enhanceSkill
- video-motion-transferSkill
- video-multimodal-generateSkill
- video-quality-enhanceSkill
- video-resolution-upscaleSkill
- video-stitchSkill
- video-to-gifSkill
The file
Files
图片换脸(image-face-swap)
Overview
双图换脸/换头合成,将源图人脸替换到目标场景图上。保留目标图的构图/身体/背景,自动肤色融合光影匹配;用于角色扮演、把自己的脸放到另一张图上、趣味合成。一次只换一张脸。
执行前应让用户清楚知道:本 Skill 会读取 Meitu 凭证、调用本地 meitu CLI、将两张人脸相关图片发送到 Meitu OpenAPI 处理,并把结果写入 ./output/ 或 $VISUAL/output/image-face-swap/。
API Mapping
- 双图换脸/换头合成:
image_face_swap
Dependencies
- meitu-cli:
>=2.0.6(npm install -g meitu-cli@latest) - 凭证:CONFIG AKSK →
meitu tools update;EXEC AKSK → 实际执行(见根CONFIG.md) - 环境变量:
MEITU_OPENAPI_TOOL_TASK_MODE=command
路径别名:
$VISUAL={OPENCLAW_HOME}/workspace/visual/
Core Workflow
Preflight → Execute → DeliverPreflight
meitu --version≥ 2.0.6(否则npm install -g meitu-cli@latest)- 确认已跑过
meitu tools update(用 CONFIG AKSK) - 当前 AKSK = EXEC,且
MEITU_OPENAPI_TOOL_TASK_MODE=command - 解析 output_dir:openclaw.yaml →
./output/|else →$VISUAL/output/image-face-swap/;mkdir -p
Execute
触发信号 / 路由规则
核心判断维度:图片数量 + head/sence 角色分配。
| 场景 | 判定 | 路由 |
| 双图换脸 | 提供 2 张图,意图换脸 | image_face_swap,head/sence 按规则分配 |
决策顺序:
- 图片数量:不足 2 张 → 提示补图;超过 2 张 → 取前 2 张并告知
- 角色分配:
- 用户明确指定 → 按指定分配 head / sence
- 未区分 → 第 1 张 = head(人脸源),第 2 张 = sence(场景底图)
- "用 B 换到 A" → B=head(源),A=sence
- 唯一路由:调用
image_face_swap
单 API 工具:参数齐全即调用,失败重试 1 次后返回错误;不做跨工具兜底。侧脸/遮挡严重不阻断调用,但告知效果可能受限。
参数定义
| 参数 | 类型 | 必填 | 默认值 | 说明 |
head_image_url | STRING | 是 | -- | 人脸源图 URL |
sence_image_url | STRING | 是 | -- | 场景底图 URL(保留构图/身体/背景) |
prompt | STRING | 是 | -- | 附加提示词。缺失 → 提示"请描述换脸效果要求" |
工具调用
meitu image-face-swap \
--skill_name skill_image-face-swap \
--head_image_url <head_url> \
--sence_image_url <sence_url> \
--prompt "<effect_description>" \
--json \
--download-dir {output_dir}错误降级
| 场景 | 处理方式 |
| 图片不足 2 张 | 提示"请提供两张图片:一张人脸源图,一张目标场景图" |
| 图片 3+ 张 | 取前 2 张(第 1=head,第 2=sence),告知用户 |
head_image_url / sence_image_url 不可访问 | 返回图片链接无效错误,不重试 |
| 未检测到人脸(head) | 返回错误:人脸源图需包含清晰正脸 |
image_face_swap 调用失败 | 重试 1 次,仍失败返回错误 |
| 内容合规拦截 | 返回合规提示,不重试 |
| 侧脸 / 遮挡严重 | 不阻断调用,告知效果可能受限 |
| 视频换脸 | 拒绝,仅支持图片 |
| 多人同时换脸 | 本工具一次只换一张脸 |
Deliver
- 直接使用 Preflight 解析的 output_dir
- 从
downloaded_files[0].saved_path读取已下载文件路径 mv {downloaded_files[0].saved_path} {output_dir}/{YYYY-MM-DD}_{descriptive}_image-face-swap.jpg
Output
- 格式:JPEG/PNG(保持原格式)
- 命名:
{YYYY-MM-DD}_{descriptive}_image-face-swap.jpg - 位置:项目 →
./output/,一次性 →$VISUAL/output/image-face-swap/
基线 Task ID
见 references/task-id-baseline.md 中对应行。
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

