ffmpeg

FFmpeg Skill

Process video and audio files using FFmpeg running inside a Docker container. Convert formats, extract audio, resize videos, create thumbnails, compress media, and more - all without installing FFmpeg locally.

Overview

This skill provides access to FFmpeg's powerful media processing capabilities through a containerized runtime. All processing happens inside an isolated Docker container with resource limits and no network access, ensuring security while maintaining full FFmpeg functionality.

When to Use This Skill

Use this skill when you need to:

• Convert video/audio between formats (MP4, WebM, AVI, MKV, MP3, WAV, etc.)
• Extract audio from video files
• Resize or scale videos to different resolutions
• Compress videos to reduce file size
• Create thumbnails from videos
• Trim or cut video clips
• Convert videos to animated GIFs
• Merge multiple video files
• Add or extract subtitles
• Adjust video bitrate and quality

Prerequisites

Docker

FFmpeg runs inside a Docker container, so Docker must be installed:

# macOS (using Homebrew)
brew install --cask docker

# Or download from: https://www.docker.com/products/docker-desktop

# Linux (Ubuntu/Debian)
sudo apt-get update
sudo apt-get install docker.io

# Start Docker service
sudo systemctl start docker
sudo systemctl enable docker

Verify Docker Installation

docker --version
docker ps  # Should list running containers

Configuration

Add to your .skill-engine.toml:

[skills.ffmpeg]
source = "docker:linuxserver/ffmpeg:latest"
runtime = "docker"
description = "FFmpeg video/audio processing"

[skills.ffmpeg.docker]
image = "linuxserver/ffmpeg:latest"
entrypoint = "ffmpeg"
volumes = ["${PWD}:/workdir"]
working_dir = "/workdir"
memory = "2g"
cpus = "4"
network = "none"
rm = true

Configuration Explained

• image: linuxserver/ffmpeg:latest - Official FFmpeg Docker image (~200MB)
• entrypoint: ffmpeg - Directly runs FFmpeg command
• volumes: ${PWD}:/workdir - Mounts current directory as /workdir in container
• working_dir: /workdir - Sets container working directory
• memory: 2g - Limits container to 2GB RAM
• cpus: 4 - Limits to 4 CPU cores
• network: none - Disables network access (processes local files only)
• rm: true - Automatically removes container after execution