How to install video-to-gif
npx skills add https://github.com/zc277584121/marketing-skills --skill video-to-gifFull instructions (SKILL.md)
Source of truth, from zc277584121/marketing-skills.
name: video-to-gif description: Convert a video to multiple GIF variants with different quality/size tradeoffs. Generates a comparison set so the user can visually pick the best result.
Skill: Video to GIF
Convert a video file into multiple GIF variants with different parameters, so the user can visually compare and pick the best one.
Prerequisite: FFmpeg and uv must be installed. gifsicle is optional (enables lossy compression variants).
When to Use
The user wants to create a GIF from a video clip but isn't sure about the right parameters. GIF quality involves tradeoffs between:
- File size — smaller is better for sharing/embedding
- Color accuracy — fewer colors = smaller but may cause banding
- Smoothness — higher FPS = smoother but larger
- Resolution — wider = sharper detail but larger
Rather than guessing, this skill generates multiple variants and lets the user decide.
Default Workflow
When the user provides a video file:
uv run --python 3.12 /path/to/skills/video-to-gif/scripts/video_to_gif.py <input.mp4>
This generates GIFs in <input>_gifs/ directory with the full preset (18 variants):
- 3 FPS options: 10, 15, 20
- 3 widths: 480px, 640px, 800px
- 2 color counts: 128, 256
Output includes a sorted comparison table showing file size, FPS, width, and colors for each variant.
Presets
| Preset | Variants | Best For |
|---|---|---|
full | ~18 | General use — broad exploration of the parameter space |
minimal | ~4 | Quick comparison — just a few key tradeoff points |
lossy | ~12 | Smallest files — includes gifsicle lossy compression levels |
quality | ~12 | Best visuals — higher res, includes bayer dithering |
# Quick comparison with fewer variants
uv run --python 3.12 .../video_to_gif.py input.mp4 --presets minimal
# Include lossy compression (requires gifsicle)
uv run --python 3.12 .../video_to_gif.py input.mp4 --presets lossy
# Higher quality focus
uv run --python 3.12 .../video_to_gif.py input.mp4 --presets quality
Common Options
| Flag | Default | Description |
|---|---|---|
-o, --output-dir | <input>_gifs/ | Output directory for all GIF variants |
--start | none | Start time in seconds (trim source) |
--end | none | End time in seconds (trim source) |
--presets | full | Preset config: full, minimal, lossy, quality |
--fps | preset | Override FPS values (e.g., --fps 10 15 20) |
--width | preset | Override width values (e.g., --width 480 640) |
--colors | preset | Override color counts (e.g., --colors 128 256) |
--lossy | preset | Gifsicle lossy levels (e.g., --lossy 0 30 80) |
Examples
# Convert first 10 seconds of a video
uv run --python 3.12 .../video_to_gif.py demo.mp4 --end 10
# Extract a specific segment
uv run --python 3.12 .../video_to_gif.py demo.mp4 --start 5 --end 15
# Custom parameter sweep
uv run --python 3.12 .../video_to_gif.py demo.mp4 --fps 12 15 --width 480 800 --colors 256
# Lossy compression comparison (needs gifsicle)
uv run --python 3.12 .../video_to_gif.py demo.mp4 --lossy 0 30 60 100
How to Choose
After running, open the output directory and compare:
- Start with the smallest files — check if quality is acceptable
- Look for color banding — if visible, try 256 colors or bayer dithering (quality preset)
- Check smoothness — if too choppy, go up to 15 or 20 FPS
- Check clarity — if text is unreadable, go up to 640 or 800px width
The sweet spot for most screen recordings is usually around 640px, 15fps, 256 colors.
Important Notes
- Widths larger than the source video resolution are automatically skipped.
- The script uses FFmpeg's two-pass palette generation for optimal GIF quality (much better than single-pass).
- Lossy compression via gifsicle can reduce file size by 30-70% with minimal visual impact at level 30-60.
- For very long clips, consider trimming with
--start/--endfirst — GIFs over 10 seconds can get very large.
Related skills
More from zc277584121/marketing-skills and the wider catalog.
remove-ai-style
Review and adjust writing style to reduce AI-generated patterns, making text read more naturally and human-like. Supports Chinese and English.
browser-screenshot
Take focused, region-specific screenshots from web pages. Navigates to the right page based on user context (URL, search query, social media post), locates the target region via DOM selectors, and crops to a clean, focused screenshot.
content-rewrite
Adapt and rewrite content for different platforms (LinkedIn, X, Reddit, English blog, WeChat). Each platform has its own tone, format, and length requirements.
chrome-automation
Automate Chrome browser tasks using agent-browser CLI. Navigate pages, fill forms, click buttons, take screenshots, extract data, and replay recorded workflows — all inside the user's real Chrome session.
image-generation
Generate illustration images for articles and documentation with a Codex-first workflow, OpenAI API fallback, and Gemini fallback.
mermaid-to-image
Convert Mermaid code blocks in Markdown files to PNG images using the mermaid.ink API.