← 返回 Skills 市场

图片水印工具

Name: 图片水印工具
Author: freedompixels

作者 freedompixels · GitHub ↗ · v1.1.0 · MIT-0

cross-platform ✓ 安全检测通过

129

总下载

当前安装

版本数

在 OpenClaw 中安装

/install cn-image-watermark

功能描述

图片水印工具。为图片添加文字或Logo水印，防止内容被搬运。支持批量处理、九宫格定位、透明度调节。当用户说"水印"、"图片水印"、"添加水印"、"防伪"、"版权"时触发。 Keywords: 图片水印, 添加水印, 防搬运, 版权保护, watermark, image.

使用说明 (SKILL.md)

cn-image-watermark - 图片水印工具

为图片添加文字或图片水印，防止搬运。

核心功能

文字水印：叠加文字（半透明、位置可调）
图片水印：叠加Logo/小图标
批量处理：对整个文件夹批量加水印
位置控制：九宫格位置（9个锚点）
透明度：0-100可调

使用场景

知识付费内容防搬运（水印"，禁止转载"）
社交媒体图片品牌标识
封面图添加版权信息
批量处理产品图片

输出格式

{
  "input": "photo.jpg",
  "output": "photo_watermarked.jpg",
  "type": "text",
  "position": "右下",
  "status": "ok"
}

使用方式

# 文字水印
python ~/.qclaw/skills/cn-image-watermark/watermark.py text "photo.jpg" "©养虾记" --position bottom-right

# Logo水印
python ~/.qclaw/skills/cn-image-watermark/watermark.py image "photo.jpg" "logo.png" --position bottom-right --opacity 60

# 批量处理
python ~/.qclaw/skills/cn-image-watermark/watermark.py batch "./photos" "output" "©养虾记"

依赖

Python3
Pillow（PIL）：pip3 install Pillow

This skill appears to do exactly what it says: add text or image watermarks to local images and folders, and it does not phone home or require credentials. Before installing/using: (1) ensure Pillow (PIL) is installed; (2) test on a small set of images — batch mode has a bug that can crash the script (exit logic expects a 'status' field that batch responses don't include); (3) adjust font paths or install a suitable TrueType font on non-macOS systems if Chinese text rendering is important; (4) run it only on directories you intend to modify (it will create output files and may overwrite if you choose output paths); and (5) review and run the code locally if you want to be certain it only touches intended files. If you need help patching the batch exit bug or making fonts portable, ask for a small patch recommendation.

功能分析

Type: OpenClaw Skill Name: cn-image-watermark Version: 1.1.0 The skill is a legitimate utility for adding text and image watermarks to photos, supporting both single-file and batch processing. The Python scripts (watermark.py and scripts/image_watermark.py) use the standard Pillow library for image manipulation and do not exhibit any high-risk behaviors such as network access, shell command execution, or data exfiltration.

能力评估

✓ Purpose & Capability

The name/description (image watermarking, batch processing, position, opacity) align with the included Python scripts. Required binaries and environment variables are none, which is consistent because the tool only needs Python and Pillow. The code operates on local image files and implements the advertised features.

ℹ Instruction Scope

SKILL.md and the scripts only instruct local file operations (reading images, writing watermarked outputs). No instructions to collect or transmit external data. Notes: (1) batch mode in watermark.py has a logic bug — the final exit uses result['status'] which doesn't exist for batch responses and will raise an exception; (2) both scripts assume macOS system font paths (PingFang etc.), which may not exist on Linux/Windows and will fall back to default fonts; (3) batch processing will recursively read image files under the provided input directory — make sure you don't point it at sensitive system folders or huge trees.

✓ Install Mechanism

There is no install spec (instruction-only / bundled scripts). That is lower risk: nothing is downloaded or executed automatically beyond the included code. The only declared dependency is Pillow (pip install), which is reasonable for image processing.

✓ Credentials

The skill requests no environment variables, no credentials, and no config paths. This is proportionate for an image watermarking utility that only reads and writes local files.

✓ Persistence & Privilege

The skill is not forced-always, does not request elevated privileges, and does not modify other skills or system configs. Its filesystem writes are limited to the output paths provided by the user/arguments.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install cn-image-watermark
安装完成后，直接呼叫该 Skill 的名称或使用 /cn-image-watermark 触发
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

v1.1.0

v1.1.0: 修复SKILL.md格式

v1.0.1

- Keywords in the description updated for better search and relevance. - English keywords replaced with more descriptive Chinese terms such as "图片水印", "添加水印", "防搬运", and "版权保护".

v1.0.0

- Initial release of cn-image-watermark. - 支持为图片添加文字或Logo水印。 - 提供九宫格水印位置选择与透明度调节。 - 支持批量处理图片文件夹。 - 通过关键词如“水印”、“防伪”自动触发使用。

元数据

Slug cn-image-watermark

版本 1.1.0

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 3

常见问题