Skip to content

Claude Code Media Generator

AI-Powered Video & Image Generation

Create stunning videos and images using Google Veo & Imagen APIs

Why Claude Code Media Generator?

  • 8 Video Generation Modes

    From text-to-video to advanced video extension, style transfer, and object manipulation. All powered by Google Veo 3.1.

    Explore all modes

  • Reference Image System

    Use up to 3 reference images for subject preservation or apply style transfer from any image.

    Learn more

  • Cloud-Native Architecture

    Direct GCS storage integration. Auto-upload for seamless video extension workflows.

    GCS Setup Guide

  • Smart Presets

    Quick, Quality, Extend, Budget — choose the right balance of speed, quality, and cost.

    View presets

Video Generation Modes

  • Text-to-Video

    Generate video from text prompt only

  • Image-to-Video

    Animate a single image with AI

  • First & Last Frames

    Interpolate between two keyframes

  • Video Extension

    Add 7 more seconds to any video

  • Reference Asset

    Preserve subject across generations

  • Reference Style

    Apply artistic style from an image

  • Insert Objects

    Add objects using mask (Veo 3.1)

  • Remove Objects

    Remove objects using mask (Veo 3.1)

View all 8 modes in detail

Quick Start

The easiest way - use the /generative skill in Claude Code:

# Generate an image
/generative image "A beautiful sunset over mountains"

# Generate a video
/generative video "Ocean waves crashing on the beach"

# With options
/generative video "Cinematic sunset" --preset quality

Learn more about Skills

For automation and advanced users:

# Option 1: Environment variable
export GOOGLE_API_KEY="your-api-key"

# Generate video
python video_gen.py "A cat playing in a sunlit garden"

# With quality preset
python video_gen.py "Cinematic sunset over mountains" --preset quality

Advanced Usage

Important: GCS Bucket Naming

Mandatory Naming Convention

When using Vertex AI API Key, your GCS bucket MUST follow this exact pattern:

gs://YOUR_PROJECT_ID-media-output/

Using any other bucket name will result in permission denied errors. This is a security boundary enforced by Vertex AI.

Example:

# ✅ Correct pattern
--storage-uri "gs://gen-lang-client-0344941103-media-output/videos/"

# ❌ Wrong - will fail
--storage-uri "gs://my-custom-bucket/videos/"

Full GCS Setup Guide

Pricing Overview

Model Type Price/Second 5-sec Video
veo-3.1-generate-preview Standard $0.35 $1.75
veo-3.1-fast-generate-preview Fast $0.15 $0.75
veo-3.0-generate-preview Previous Gen $0.25 $1.25
veo-2.0-generate-preview Legacy $0.35 $1.75

Cost Saving Tip

Use --preset budget to automatically select the most cost-effective model for your use case.

Project Status

Component Version Status
video_gen.py v2.27 Production Ready
image_gen.py v1.4 Stable
check_api.py v2.1 Stable
Documentation v3.10 Updated
Claude Code Skill v3.2.1 Available
Agent v1.0 Available
MCP Server — Planned

Development Roadmap

  • Phase 1: Core Features ✅

    • 8 video generation modes
    • Image generation (Imagen/Gemini)
    • Smart validation & auto-correction
    • GCS storage integration

  • Phase 2: Enhanced Workflows ✅

    • Video extension from URLs
    • Auto FPS conversion (24fps)
    • Reference image system
    • Smart defaults from metadata

  • Phase 3: Claude Code ✅

    • /generative unified skill
    • Help system (5 topics)
    • Config wizard
    • Agent for auto-detection

  • Phase 4: Planned Features

    • Remix Mode (--remix video.mp4)
    • Outpainting (--outpaint left,right)
    • Camera Control (--camera pan-left)
    • Audio Control (style/mood)

Explore Documentation

  • Using Skills

    The easiest way to generate media - /generative skill for Claude Code.

    Start with Skills

  • Getting Started

    Installation, authentication setup, and creating your first video in minutes.

    Get Started

  • Video Generation

    All 8 modes, model comparison, reference images, and advanced features.

    Video docs

  • Image Generation

    Text-to-image with Gemini & Imagen models, aspect ratios, and templates.

    Image docs

  • Advanced Usage

    Direct scripts, automation, CI/CD integration for power users.

    Advanced docs

  • Guides

    GCS storage, presets, troubleshooting, and more.

    View guides

Made with using Google Vertex AI