| name | download_image |
| description | A robust, "Honey Badger" image downloader that bypasses strict CDNs, ignores misleading Content-Types, and falls back to scraping if a direct link fails. |
| dependencies | {"python":["requests","beautifulsoup4"],"node":[],"binaries":[]} |
Image Downloader (Honey Badger Edition) 🖼️
This skill is designed to be resilient. It doesn't trust HTTP headers; it trusts the actual file bytes. Use this to download images from the web even when servers try to block bots or serve incorrect MIME types.
🛡️ Robust Features:
- Byte Sniffing: Ignores
Content-Type headers (often binary/octet-stream) and checks the file signature (magic numbers) to detect JPEGs, PNGs, GIFs, and WEBPs.
- Smart Fallback: If the URL returns HTML instead of an image, it automatically switches to scraping mode to find the high-res
og:image or twitter:image tags.
- Stealth Mode: Uses a modern Chrome User-Agent to bypass basic anti-bot protections.
🚀 Execution Command:
run_command("python skills/download_image/main.py '<url>' 'temp/<filename>'")
📋 Requirements:
- URL: A direct image link OR a page URL (Reddit, Pinterest, 4KWallpapers, etc.).
- Filename: Descriptive name (e.g.,
lisa_figaro.jpg). Always save to temp/.
📤 Output Handling:
To show the image to the user, use a concise description: .
[!WARNING]
DO NOT wrap your entire response inside the ![alt-text] part of the image tag. Keep the alt-text short (e.g., "image" or "Lisa photo").
⚠️ Dependencies:
