agentleFS
meitumeitu/meitu-skills/skills/image-background-replace/SKILL.md

image-background-replace

Skill33 starsChanged 5 months agoMITView on GitHub

替换图片整体背景/换场景,保留前景主体不变。当用户说换背景、替换背景、换成 xx 背景、商品换场景、人像换背景、证件照换底色、换底时触发。本工具不能输出透明背景。执行时会使用本地 meitu CLI、Meitu OpenAPI 凭证并写入本地输出目录。

  • Reads credentials
  • Installs packages

The file

Files
SKILL.md

图片换背景(image-background-replace)

Overview

对已有图片做整体背景替换/换场景,保留前景主体不变。覆盖商品换场景、人像换背景、宠物换场景、电商白底图、证件照换底色(白/蓝/红)、文字描述任意目标场景的 AI 生成。本工具不能输出透明背景。

API Mapping

  • 背景替换:image_gummy_generate_v45

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 → Deliver

Preflight

  1. meitu --version ≥ 2.0.6(否则 npm install -g meitu-cli@latest)
  2. 确认已跑过 meitu tools update(用 CONFIG AKSK)
  3. 当前 AKSK = EXEC,且 MEITU_OPENAPI_TOOL_TASK_MODE=command
  4. 解析 output_dir:openclaw.yaml → ./output/ |else → $VISUAL/output/image-background-replace/;mkdir -p

Execute

触发信号 / 路由规则

场景判定关键词路由
整体背景替换换背景、换场景、换底色、证件照换底image_gummy_generate_v45

决策顺序: 0. 前置拦截:prompt 含"透明"/"透明底"/"透明背景" → 不调用本工具,提示走抠图

  1. 参数齐全判断:image_url 缺失 → 追问;prompt 缺失 → 用默认"白色背景"
  2. 唯一路由:调用 image_gummy_generate_v45

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。

参数定义

参数类型必填范围默认值说明
image_urlSTRING是----图片地址。缺失 → 提示"请提供需要换背景的图片"
promptSTRING是--白色背景目标背景描述。用户未指定 → 默认"白色背景"并告知可描述具体场景

工具调用

meitu image-background-replace \
  --skill_name skill_image-background-replace \
  --image_url <image_url> \
  --prompt "<target_background_description>" \
  --json \
  --download-dir {output_dir}

错误降级

场景处理方式
image_url 缺失提示"请提供需要换背景的图片",不调用 API
image_url 不可访问直接返回图片链接无效错误,不重试
prompt 缺失使用默认值"白色背景"
未检测到前景主体返回错误,提示需包含可识别的前景主体,不重试
image_gummy_generate_v45 调用失败重试 1 次,仍失败返回错误
内容合规拦截直接返回合规提示,不重试
生成背景与前景不协调建议用户调整 prompt 后重试
用户同时说"换背景"和"透明底"本工具不能输出透明背景,不调用
视频换背景拒绝,本工具不支持视频

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-background-replace.png

Output

  • 格式:PNG(非透明)
  • 命名:{YYYY-MM-DD}_{descriptive}_image-background-replace.png
  • 位置:项目 → ./output/,一次性 → $VISUAL/output/image-background-replace/

基线 Task ID

见 references/task-id-baseline.md 中对应行。

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.