Token导航 LogoToken导航TokenDH.com
图像处理需要联网github未标认证来源可访问clear审计提醒

image-fetcher图像获取器

Agent Skill

用于辅助图像生成、图片编辑、视觉素材处理或图像模型工作流。它适合让 Agent 根据文本生成图片、处理背景、整理视觉提示词或调用相关图像工具。使用时需要确认输入图片、版权来源、输出格式和模型限制;涉及人物、品牌、商品或公开展示素材时,应额外核对授权、真实性和内容合规边界。

总安装

392

周安装

16

GitHub Stars

14

下载量

127
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

3

许可证

MIT

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:image-fetcher(图像获取器)
来源仓库:https://github.com/interstellar-code/claud-skills
仓库路径:skills/image-fetcher
安装命令:
npx skills add https://github.com/interstellar-code/claud-skills --skill image-fetcher
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/interstellar-code/claud-skills --skill image-fetcher

简介

从网络或本地源获取图像资源并标准化存储。

  • 适用于数据整理、素材库构建和自动化工作流。
  • 支持多种来源协议与格式自动识别。image-fetcher 属于图像处理类 Skill,可作为该场景下的辅助能力补充。
  • 需遵守各平台抓取频率限制与使用条款。适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。
  • 敏感内容下载前应确认合法性与使用权限。

SKILL.md

Image Fetcher Skill

This skill enables Claude to fetch and download images from the internet with support for multiple image formats and batch downloading capabilities.

When to Use This Skill

Use this skill when the user requests:

  • Downloading images from specific URLs
  • Fetching images from the web to embed in documents
  • Saving images locally from internet sources
  • Batch downloading multiple images
  • Converting or processing images that need to be downloaded first

Supported Image Formats

The skill automatically detects and handles these formats:

  • JPG/JPEG
  • PNG
  • GIF
  • WebP
  • BMP
  • SVG
  • ICO
  • TIFF/TIF

Core Workflows

Single Image Download

To download a single image from a URL:

python scripts/fetch_image.py <image_url> [output_directory] [filename]

Parameters:

  • image_url (required): URL of the image to download
  • output_directory (optional): Directory to save the image (defaults to current directory)
  • filename (optional): Custom filename for the saved image (defaults to URL filename)

Examples:

# Basic download to current directory
python scripts/fetch_image.py https://example.com/photo.jpg

# Download to specific directory
python scripts/fetch_image.py https://example.com/photo.jpg ./downloads

# Download with custom filename
python scripts/fetch_image.py https://example.com/photo.jpg ./downloads myimage.jpg

Batch Image Download

To download multiple images from a list of URLs:

python scripts/fetch_images_batch.py <urls_file> [output_directory]

Input file formats:

  1. Text file with one URL per line
  2. JSON file with array of URLs: ["url1", "url2",...]

Examples:

# Download from text file
python scripts/fetch_images_batch.py urls.txt

# Download to specific directory
python scripts/fetch_images_batch.py urls.json ./images

# Create a URL list on the fly
echo -e "https://example.com/img1.jpg\nhttps://example.com/img2.png" > urls.txt
python scripts/fetch_images_batch.py urls.txt ./downloads

The batch script creates a fetch_results.json file with detailed results for each download.

Integration with Claude Workflow

For Document Embedding

When fetching images to embed in documents (DOCX, PPTX, etc.):

  1. Download the image to a temporary location
  2. Use the returned file path to embed the image in the document
  3. Clean up temporary files if needed
# Download image
python scripts/fetch_image.py https://example.com/chart.png /home/claude/temp

# The script outputs the full path which can be used for embedding
# Example output: /home/claude/temp/chart.png

For Chat Display

When downloading images to display in chat:

  1. Download to /mnt/user-data/outputs/ directory
  2. Provide the user with a link to the downloaded image
python scripts/fetch_image.py https://example.com/image.jpg /mnt/user-data/outputs

Search Integration

When combined with web search:

  1. Search for images using web_search tool
  2. Extract image URLs from search results
  3. Use this skill to download the images

Error Handling

The scripts handle common scenarios:

  • Invalid URLs
  • Network timeouts
  • Unsupported formats
  • Permission errors
  • Missing directories (auto-created)

Failed downloads are reported with clear error messages.

Output Information

Each successful download provides:

  • Full path to saved file
  • File size in KB
  • Content-Type from server
  • Success confirmation

Best Practices

  1. Always validate URLs before attempting download
  2. Use batch downloading for multiple images to improve efficiency
  3. Specify output directory to organize downloads properly
  4. Check file extensions - the script auto-detects format from headers if URL lacks extension
  5. Handle errors gracefully - inform users of any failed downloads

Common Use Cases

Use Case 1: Embedding web images in presentations

# User: "Add this chart to my presentation: https://data.com/chart.png"
python scripts/fetch_image.py https://data.com/chart.png /home/claude/temp
# Then use the path to embed in PPTX

Use Case 2: Creating image gallery from URLs

# User: "Download these product images and save them"
# Create urls.txt with image URLs
python scripts/fetch_images_batch.py urls.txt /mnt/user-data/outputs/gallery

Use Case 3: Logo download for branding

# User: "Get the company logo from their website"
python scripts/fetch_image.py https://company.com/logo.svg /home/claude/assets logo.svg

Technical Notes

  • Uses requests library with appropriate headers to avoid being blocked
  • Streams large files to handle memory efficiently
  • 30-second timeout for each download
  • Automatic content-type detection and extension mapping
  • Creates output directories automatically if they don't exist

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

04

需要参考平台分布和安装热度时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

OpenCode

29.04%
按下载量换算37

Antigravity

22.07%
按下载量换算28

Claude Code

17.56%
按下载量换算22

Codex

11.98%
按下载量换算15

windsurf

8.13%
按下载量换算10

Gemini CLI

3.41%
按下载量换算4

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。

来源信息

继续浏览同类 Skills