media-preview

Extract JPEG thumbnails and metadata from media files using ffprobe and Pillow.

1|Updated Feb 14, 2026
One-click install
npx skills add https://github.com/FatOtter/project-jupiter --skill media-preview
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: media-preview
Source: https://github.com/FatOtter/project-jupiter/tree/main/.opencode/skills/media-preview
Command: npx skills add https://github.com/FatOtter/project-jupiter --skill media-preview

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires Pillow, ffprobe, and includes scripts (resource) components.

What problem does it solve?

This skill reduces context load by generating lightweight thumbnails and concise metadata for media files, allowing quick content understanding without loading full binaries into context.

Core Features & Use Cases

  • Generate JPEG thumbnails for images and video frames
  • Extract basic metadata (format, dimensions, duration, bitrate) using ffprobe and Pillow
  • Output a structured JSON summary and provide read-ready paths for thumbnails via the Read tool
  • Use Case: Quickly inspect a large collection of media before archiving or editing

Quick Start

Use the media-preview skill on a media file to obtain a thumbnail and metadata without loading the full file.

Frequently Asked Questions about media-preview

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I extract thumbnails and metadata from media files without loading full binaries into context?

To extract thumbnails and metadata without loading full binaries, use ffprobe and Pillow to parse media data and output a lightweight JSON summary with JPEG thumbnail paths for downstream tools.

Can I generate video frame thumbnails and extract duration metadata using ffprobe?

Yes, you can generate JPEG thumbnails for video frames and extract metadata including format, dimensions, duration, and bitrate by applying ffprobe to parse the media file data.

Does media-preview work with images, video, and audio in common formats?

Media-preview works with images, video, and audio in common formats, using Pillow for image thumbnails and ffprobe to parse media data across multiple file types.

What's the best way to inspect a large collection of media files before archiving?

The best way to inspect a large collection of media files before archiving is generating lightweight thumbnails and concise metadata summaries to reduce context bloat and provide quick content understanding.

Do I need to install Pillow and ffprobe to use this media inspection skill?

Yes, you need to install Pillow and ffprobe as dependencies, since they provide the core functionality for parsing image data and extracting media metadata respectively.

Why does loading full media files cause context bloat and how can metadata summaries help?

Loading full media files causes context bloat because raw binaries consume excessive space, whereas extracting concise metadata and thumbnail paths provides quick content understanding without loading full files.