JSPM

  • ESM via JSPM
  • ES Module Entrypoint
  • Export Map
  • Keywords
  • License
  • Repository URL
  • TypeScript Types
  • README
  • Created
  • Published
  • Downloads 56
  • Score
    100M100P100Q46981F
  • License MIT

AI media toolkit — give any agent the ability to create professional media

Package Exports

    This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (gunni) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

    Readme

    Gunni

    AI media toolkit. Give any AI agent the ability to create professional media.

    One CLI. Seven tools. Zero config.

    Install

    npm install -g gunni

    Quick start

    # Set up your API key
    gunni config
    
    # Generate an image
    gunni image "coffee bag product shot, warm editorial light" -o bag.png
    
    # Remove background
    gunni image bag.png --remove-bg -o bag-clean.png
    
    # Edit: place on a scene
    gunni image bag-clean.png "on marble counter, morning light" -o hero.png
    
    # Upscale
    gunni image hero.png --upscale -o hero-final.png
    
    # Describe (for alt text, iteration)
    gunni image hero-final.png
    
    # Generate video from image
    gunni video hero.png -p "slow cinematic zoom in" -o hero.mp4
    
    # Text to speech
    gunni audio "Welcome to our roastery" -o intro.mp3

    Use as an MCP server

    Connect to the Gunni API server:

    {
      "mcpServers": {
        "gunni": {
          "type": "streamable-http",
          "url": "https://gunni-api.fly.dev/mcp",
          "headers": {
            "Authorization": "Bearer YOUR_GUNNI_API_KEY"
          }
        }
      }
    }

    Commands

    Command What it does
    gunni image Generate, edit, describe, upscale, remove background (smart routing)
    gunni video Image or text to video
    gunni audio Text to speech
    gunni learn Creative expertise and best practices
    gunni list models Discover available models
    gunni config API key setup
    gunni init Start a new project with BRIEF.md
    gunni history Search past work

    How it works

    One command handles all image operations. Gunni detects what you want from the inputs:

    • Text only: generate
    • Image + text: edit
    • Image only: describe
    • --upscale: upscale
    • --remove-bg: remove background

    Models

    Curated set of the best model for each task. You don't pick models. Gunni routes to the right one.

    Capability Model
    Image generation Nano Banana (Google)
    Image editing Nano Banana Edit
    Design + typography Recraft v4
    Upscaling Topaz
    Background removal Bria
    Video Kling v3 Pro
    Text to speech MiniMax Speech HD

    Override with --model <id> when you have a reason.

    Project system

    Create a BRIEF.md in your project root to give Gunni creative direction:

    # Kova Coffee
    
    ## Colors
    - Primary: #2C1810 (espresso)
    - Accent: #D4A574 (caramel)
    
    ## Style
    - Warm editorial photography, natural light
    - Minimalist, lots of negative space

    Run gunni init --template brand to scaffold a project.

    License

    MIT