← 返回 Skills 市场

Image Tools (ImageMagick)

Name: Image Tools (ImageMagick)
Author: pratiknarola

作者 Pratik Narola · GitHub ↗ · v1.0.0 · MIT-0

linuxdarwin ⚠ suspicious

总下载

当前安装

版本数

在 OpenClaw 中安装

/install image-tools-magick

功能描述

Comprehensive image editing toolkit using ImageMagick. Resize, crop, composite (overlay), pad, annotate, adjust (brightness/contrast/blur/rotate/flip), remov...

使用说明 (SKILL.md)

Image Tools (ImageMagick)

Pixel-level image manipulation via shell scripts. For AI-based editing (add/remove content with prompts), see nano-banana-pro.

Requires: imagemagick (apt install imagemagick / brew install imagemagick)

Scripts

All scripts are in scripts/ relative to this skill. All output to a new file (non-destructive).

🔧 resize.sh — Resize images

scripts/resize.sh \x3Cinput> \x3Cgeometry> [output]

Geometry	Effect
`800x`	Width 800, keep aspect ratio
`800x600`	Fit within 800x600
`800x600!`	Force exact 800x600 (distort)
`50%`	Scale to 50%
`800x800\>`	Shrink only if larger

✂️ crop.sh — Crop images

scripts/crop.sh \x3Cinput> \x3CWxH+X+Y> [output]

500x500+100+50 — crop 500x500 starting at pixel (100, 50)
500x500+center — center crop (special mode)

🧩 composite.sh — Overlay / place image on image

scripts/composite.sh \x3Cbackground> \x3Coverlay> [output] [options]

Options: --gravity, --offset +X+Y, --resize GEO, --opacity PCT

Place logo on photo: scripts/composite.sh bg.jpg logo.png out.jpg --gravity southeast
Watermark: scripts/composite.sh bg.jpg mark.png out.jpg --opacity 30 --resize 200x200

📐 pad.sh — Add padding / extend canvas

scripts/pad.sh \x3Cinput> \x3CWxH> [output] [--color COL] [--gravity POS]

Make square: scripts/pad.sh wide.jpg 1080x1080 --color white
Transparent pad: scripts/pad.sh icon.png 512x512 --color none

🎨 adjust.sh — Brightness, contrast, rotate, flip, blur, etc.

scripts/adjust.sh \x3Cinput> [output] [options]

Options: --brightness N, --contrast N, --saturation N, --rotate N, --flip, --flop, --grayscale, --blur 0xN, --sharpen 0xN, --negate, --border WxH, --border-color COL

🔤 annotate.sh — Add text overlay

scripts/annotate.sh \x3Cinput> \x3Ctext> [output] [options]

Options: --font, --size, --color, --bg, --gravity, --offset, --stroke, --stroke-width

🧹 remove-bg.sh — Remove solid background color → transparent

scripts/remove-bg.sh \x3Cinput> \x3Coutput> [tolerance%] [color]

Remove white: scripts/remove-bg.sh icon.png clean.png
Remove green screen: scripts/remove-bg.sh photo.png clean.png 25 "#00FF00"

🔄 convert-format.sh — Format conversion

scripts/convert-format.sh \x3Cinput> \x3Coutput> [--quality N] [--strip]

Supports: PNG, JPG, WebP, GIF, BMP, TIFF

ℹ️ info.sh — Image metadata

scripts/info.sh \x3Cimage>

Direct ImageMagick (for anything not covered by scripts)

# Tile/montage multiple images
montage img1.jpg img2.jpg img3.jpg -geometry 300x300+5+5 montage.jpg

# Generate app icon set
for size in 1024 512 256 128 64 32 16; do
  convert icon.png -resize ${size}x${size} icon-${size}.png
done

# Rounded corners
convert input.png \( +clone -alpha extract -draw 'fill black polygon 0,0 0,15 15,0 fill white circle 15,15 15,0' \
  \( +clone -flip \) -compose Multiply -composite \( +clone -flop \) -compose Multiply -composite \) \
  -alpha off -compose CopyOpacity -composite rounded.png

# Append images (horizontal / vertical)
convert img1.jpg img2.jpg +append horizontal.jpg    # side by side
convert img1.jpg img2.jpg -append vertical.jpg      # stacked

AI-Based Editing

For adding/removing objects, style transfer, or content-aware edits, use the nano-banana-pro skill (Gemini image editing):

uv run /root/shared/skills/nano-banana-pro/scripts/generate_image.py \
  --prompt "remove the person on the left" \
  --input-image photo.jpg \
  --filename edited.png

安全使用建议

Review before installing. Use only with trusted inputs, and avoid passing untrusted annotation text, paths, colors, offsets, rotation, blur, sharpen, or border values until the helper scripts replace eval with safely quoted argument arrays and basic option validation.

能力标签

crypto

能力评估

ℹ Purpose & Capability

The artifacts match the stated ImageMagick image-editing purpose: resize, crop, annotate, adjust, composite, pad, convert, remove backgrounds, and inspect metadata. The main concern is implementation safety, not a mismatch in advertised capability.

⚠ Instruction Scope

The skill tells agents to pass user-chosen image parameters into shell scripts, and annotate.sh and adjust.sh then re-parse those values through eval. That gives crafted text, colors, geometry, paths, or transform options more authority than the documented image operation requires.

✓ Install Mechanism

No install-time execution, package-install hook, credential request, persistence setup, or hidden installer behavior was found in the provided artifacts or metadata.

⚠ Credentials

Local ImageMagick scripts are proportionate for this purpose, but eval-based command construction is not proportionate because it can expand a normal image-editing request into arbitrary local shell execution.

✓ Persistence & Privilege

No background workers, startup persistence, privilege escalation, credential harvesting, or network exfiltration behavior was identified. The risk is limited to invocation-time shell injection in specific helpers.

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 image-tools-magick
安装完成后，直接呼叫该 Skill 的名称或使用 /image-tools-magick 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

v1.0.0

Initial release: resize, crop, composite, pad, annotate, adjust, remove-bg, convert-format, info — full ImageMagick toolkit for AI agents

元数据

Slug image-tools-magick

版本 1.0.0

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 1

常见问题