AI media generation CLI tool using Google's Imagen 4, Veo 3.1, and Gemini TTS. Use when the user wants to (1) generate images from text prompts, (2) edit existing images with AI, (3) explain image contents, (4) generate videos from text or images, (5) create narration/voice audio with character settings. Triggers on requests like "generate an image of...", "create a video...", "make a voice that says...", "edit this image to...", "describe this image".
Standardmäßig ist der Prompt ausgewählt, der zuerst die Quelle prüft. Sie können zu einem direkten Befehl wechseln oder eine lokale Kopie herunterladen.
Quelldateien prüfen
Lesen Sie SKILL.md und alle von SkillsMP angezeigten Begleitdateien, bevor Sie sich für eine Installation entscheiden.
Mit Codex oder Claude installieren Kopieren Sie diesen Prompt, fügen Sie ihn in Codex, Claude oder einen anderen Assistant ein und lassen Sie die Skill-Seite prüfen und installieren.
Ein direkter Befehl überspringt den Prüf-Prompt. Prüfen Sie die Quelle, bevor Sie ihn ausführen.
AI media generation CLI tool using Google's Imagen 4, Veo 3.1, and Gemini TTS. Use when the user wants to (1) generate images from text prompts, (2) edit existing images with AI, (3) explain image contents, (4) generate videos from text or images, (5) create narration/voice audio with character settings. Triggers on requests like "generate an image of...", "create a video...", "make a voice that says...", "edit this image to...", "describe this image".
license
MIT
metadata
{"author":"hirokidaichi","version":"0.4.0"}
ergon - AI Media Generation CLI
Note: Run with npx ergon if not installed globally.
Quick Reference
npx ergon image gen "<prompt>" -t <style> -a <ratio> # Image generation
npx ergon image edit <file> "<instruction>"# Image editing
npx ergon video gen "<prompt>" [-i <image>] # Video with audio
npx ergon narration gen "<text>" -c "<character>"# Voice generation
npx ergon image gen "cute cat mascot for tech startup" -t anime -a 1:1
npx ergon image gen "professional team meeting in modern office" -t corporate -a 16:9
npx ergon image gen "abstract geometric logo" -t minimal -a 1:1 -o logo.png
Image Editing
npx ergon image edit [options] <file> <prompt>
Edit instructions in natural language:
Background change: "change background to sunset beach"
Style transfer: "make it look like watercolor painting"
Object removal: "remove the person on the left"
Color adjustment: "make colors more vibrant"
npx ergon image edit photo.jpg "change background to blue sky"
npx ergon image edit portrait.png "convert to anime style"
Video Generation (with Audio)
Veo 3.1 generates videos with synchronized audio. Include audio/sound instructions directly in the prompt.
npx ergon video gen [options] <theme>
Prompt Structure for Audio-Video
Include sound descriptions in your prompt:
# Sound effects included
npx ergon video gen "cat meowing and playing with a ball, soft purring sounds"# Music/ambient audio
npx ergon video gen "sunset timelapse over ocean, with calming wave sounds and soft piano music"# Dialogue/voice
npx ergon video gen "person saying 'welcome to our channel' with friendly tone, waving at camera"
Image-to-Video
Animate a static image with motion and sound:
npx ergon video gen "character starts dancing to upbeat music" -i character.png
npx ergon video gen "logo reveals with whoosh sound effect" -i logo.png
Options
Option
Values
Default
-i, --input
image file
-
-d, --duration
5-8 seconds
8
-a, --aspect-ratio
16:9, 9:16
16:9
--fast
use Veo 3.1 Fast
false
Vertical video for TikTok/Reels: -a 9:16
Narration Generation
For voice-only audio without video, use narration command.
npx ergon narration gen [options] <text>
Character and Acting Direction
Use -c (character) and -d (direction) for expressive voice:
# Character defines WHO is speaking
npx ergon narration gen "Let's go on an adventure!" -c "energetic young girl"# Direction defines HOW they speak
npx ergon narration gen "The results are in..." -c "news anchor" -d "serious, building suspense"# Combined for full expression
npx ergon narration gen "Yay! We did it!" -c "excited child" -d "jumping with joy, high energy"
Voice Selection
Voice
Character
Kore
Female, versatile (default)
Aoede
Female, warm
Charon
Male, deep
Fenrir
Male, strong
Puck
Neutral, playful
Options
Option
Values
Default
-v, --voice
Kore, Aoede, Charon, Fenrir, Puck
Kore
-c, --character
character description
-
-d, --direction
acting direction
-
--speed
0.25-4.0
1.0
-l, --lang
ja, en, zh, ko, etc.
ja
Workflow Patterns
Generate, then Edit
npx ergon image gen "product photo of headphones" -t realistic
npx ergon image edit headphones.png "add soft shadow, white background"
Image to Animated Video
npx ergon image gen "mascot character standing" -t anime -a 1:1
npx ergon video gen "mascot waves and says hello cheerfully" -i mascot.png
Preview Before Generation
npx ergon image gen "complex scene" --dry-run # Check settings
npx ergon video gen "expensive render" --dry-run # Verify before API call
