← Back to Skills Marketplace
194
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install bg-remover
Description
Remove, replace, or blur image backgrounds using AI-powered segmentation (rembg/U2-Net). Use when the user asks to: (1) remove image background / make transp...
README (SKILL.md)
bg-remover — 图片背景处理
基于 rembg (U2-Net) 的本地 AI 抠图工具，支持去除背景、替换背景、背景模糊。
依赖安装
首次使用前运行安装脚本：
bash {baseDir}/scripts/install.sh
或手动安装：pip install rembg Pillow numpy onnxruntime
命令
所有操作通过 {baseDir}/scripts/bg_remove.py 执行：
去除背景（输出透明 PNG）
python3 {baseDir}/scripts/bg_remove.py remove \x3Cinput> [-o output.png]
替换背景
# 纯色背景（支持 hex、rgb、颜色名）
python3 {baseDir}/scripts/bg_remove.py replace \x3Cinput> -c "#FF0000" [-o output.png]
python3 {baseDir}/scripts/bg_remove.py replace \x3Cinput> -c white [-o output.png]
# 图片背景
python3 {baseDir}/scripts/bg_remove.py replace \x3Cinput> -i bg.jpg [-o output.png]
背景模糊（人像模式）
python3 {baseDir}/scripts/bg_remove.py blur \x3Cinput> [-r 15] [-o output.png]
-r 控制模糊半径，值越大越模糊，默认 15。
输出
- 默认输出路径为原文件名加后缀：
_nobg.png（去除）、_newbg.png（替换）、_blur.png（模糊） - 所有输出均为 PNG 格式
- 使用
-o参数自定义输出路径
注意事项
- 首次运行会自动下载 U2-Net 模型（约 170MB），需要网络
- 处理大图可能较慢，建议先缩小再处理
- 抠图效果取决于前景与背景的对比度，简洁背景效果更好
Usage Guidance
This skill appears coherent and implements local image background removal. Before installing: (1) be aware the first run will download a ~170MB U2‑Net model and pip will fetch packages from PyPI (onnxruntime can be large); (2) run the install script inside a Python virtualenv to avoid polluting system packages; (3) review/approve network access if you are on a metered connection; (4) do not feed sensitive images unless you are comfortable with local processing and the rembg package’s behavior; and (5) if you need stricter isolation, consider running the script in a container or sandbox. Overall there are no signs of credential access or hidden exfiltration in the provided files.
Capability Analysis
Type: OpenClaw Skill
Name: bg-remover
Version: 1.0.0
The bg-remover skill bundle is a legitimate tool for image background processing. It uses well-known libraries (rembg, Pillow) to perform background removal, replacement, and blurring. The installation script (scripts/install.sh) and the main logic (scripts/bg_remove.py) are transparent, follow the stated purpose, and contain no signs of data exfiltration, malicious execution, or prompt injection.
Capability Assessment
Purpose & Capability
Name/description (background removal, replacement, blur) match the provided files: a Python CLI (scripts/bg_remove.py) that implements remove/replace/blur and an install script to install rembg and its dependencies. Required resources (rembg, Pillow, numpy, onnxruntime) are appropriate for the task.
Instruction Scope
SKILL.md instructs running the included install.sh and the bg_remove.py commands; the script only reads input image files and writes PNG outputs. The SKILL.md and script explicitly note a one-time model download (~170MB). There are no instructions to read unrelated files, environment variables, or send processed data to external endpoints.
Install Mechanism
Installation is via the included scripts/install.sh which runs pip install rembg Pillow numpy onnxruntime (PyPI). This is a standard package install from PyPI; there are no downloads from arbitrary URLs, URL shorteners, or archive extraction from unknown hosts in the repo.
Credentials
The skill declares no required environment variables, no credentials, and the code does not read any environment secrets. The requested dependencies are proportionate to the image-processing functionality.
Persistence & Privilege
The skill does not request permanent/always-on presence, does not modify other skills or system-wide configs, and is user-invocable only. It performs no privileged operations.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install bg-remover - After installation, invoke the skill by name or use
/bg-remover - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of bg-remover:
- AI-powered tool for removing, replacing, or blurring image backgrounds using rembg (U2-Net).
- Supports background removal, solid color or image background replacement, and portrait-style background blur.
- Command-line script with flexible color/image input options and adjustable blur radius.
- Outputs processed images as PNG files, with customizable output paths.
- Automatically downloads required model on first use; installation instructions provided.
Metadata
Frequently Asked Questions
What is bg-remover?
Remove, replace, or blur image backgrounds using AI-powered segmentation (rembg/U2-Net). Use when the user asks to: (1) remove image background / make transp... It is an AI Agent Skill for Claude Code / OpenClaw, with 194 downloads so far.
How do I install bg-remover?
Run "/install bg-remover" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is bg-remover free?
Yes, bg-remover is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does bg-remover support?
bg-remover is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created bg-remover?
It is built and maintained by Jianghua (@jxhgzs); the current version is v1.0.0.
More Skills