aethershell 0.3.1

The world's first multi-agent shell with typed functional pipelines and multi-modal AI
Documentation
# AI Features Demo
# Set AETHER_AI=openai and OPENAI_API_KEY for real responses

# Simple text prompt
response = ai("What is 2+2?")
print(response)

# With model specification (when using real AI)
# response = ai("Explain quantum computing", {model: "openai:gpt-4o"})

# Multi-modal: describe an image
# ai("Describe this image", {images: ["photo.jpg"]})

# Multi-modal: transcribe audio  
# ai("Transcribe this audio", {audio: ["recording.mp3"]})

# Multi-modal: analyze video
# ai("Summarize this video", {video: ["clip.mp4"]})

# Combined multi-modal
# ai("Analyze this presentation", {
#   images: ["slide1.png", "slide2.png"],
#   audio: ["narration.mp3"]
# })