Does this skill require an NVIDIA GPU?

No, a GPU is not required, but if an NVIDIA GPU and the Container Toolkit are detected, the setup will optimize the environment for faster transcription speeds.

What is the difference between the Lite and Full image variants?

The Lite variant (~500MB) includes tools for YouTube and screen recording analysis, while the Full variant (~4GB) adds Whisper for local video speech transcription.

Can I use this setup on Windows?

Yes, it is fully compatible with Windows via WSL (Windows Subsystem for Linux), including automatic path mapping for Windows screenshot folders.

Where does Claude Vision store its configuration files?

All settings and local video caches are stored in your home directory at ~/.claude/claude-vision/.

Claude Vision Setup

Name: Claude Vision Setup
Author: ellyseum

byellyseum

0•

生产力与工作流

Configures the environment to enable screenshot, clipboard, and video analysis capabilities within Claude Code.

The Claude Vision Setup skill provides an interactive wizard to bridge Claude Code with your local visual environment. It automatically detects system resources like Docker and NVIDIA GPUs to install appropriate image variants, sets up screenshot directory mappings for various OS types (WSL, macOS, Linux), and configures local transcription services. This ensures Claude can seamlessly analyze visual context from your screen, clipboard, or media files with minimal manual configuration, providing a robust foundation for multi-modal developer workflows.

主要功能

01Interactive configuration for Lite and Full Docker image variants

02Automated environment detection for Docker, GPU, and OS types

03OS-specific directory mapping for instant screenshot access

04Local video analysis setup with Whisper transcription support

050 GitHub stars

06Integrated health checks to verify container and configuration status

使用场景

01Setting up local video processing to analyze screen recordings and tutorials

02Configuring cross-platform visual context tools for WSL, macOS, and Linux

03Enabling Claude to see and debug UI issues directly from your latest screenshots

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add ellyseum/claude-vision claude-vision-setup

For use in Claude.ai and ChatGPT

主要功能

01Interactive configuration for Lite and Full Docker image variants

02Automated environment detection for Docker, GPU, and OS types

03OS-specific directory mapping for instant screenshot access

04Local video analysis setup with Whisper transcription support

050 GitHub stars

06Integrated health checks to verify container and configuration status

使用场景

01Setting up local video processing to analyze screen recordings and tutorials

02Configuring cross-platform visual context tools for WSL, macOS, and Linux

03Enabling Claude to see and debug UI issues directly from your latest screenshots

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add ellyseum/claude-vision claude-vision-setup

For use in Claude.ai and ChatGPT