Skip to content

About

Fixed version of an MCP server for generating images with MCP / Bedrock

Resources

Stars

0 stars

Watchers

0 watching

Forks

Latest commit

 

History

4 Commits

Folders and files

Repository files navigation

mcp_bedrock_media_generator

Fixed version of an MCP server for generating images with MCP / Bedrock

AWS Bedrock Media Generator MCP Server

A Model Context Protocol (MCP) server for generating images and videos using Amazon Bedrock's Nova Canvas and Nova Reel models. Designed specifically for LibreChat's MCP image generation feature.

Features

  • Image Generation: High-quality images using Amazon Nova Canvas
  • Video Generation: Video creation using Amazon Nova Reel
  • LibreChat Compatible: Returns images in proper MCP format for inline display
  • SSE Transport: Server-Sent Events for real-time communication
  • Docker Ready: Easy deployment with Docker/Docker Compose
  • AWS Integration: Direct integration with AWS Bedrock services

Quick Start

Using Docker (Recommended)

# Pull and run the container
docker run -d -p 8961:8961 \
  -e AWS_ACCESS_KEY_ID=your_access_key \
  -e AWS_SECRET_ACCESS_KEY=your_secret_key \
  -e AWS_REGION=us-east-1 \
  --name mcp-media-generator \
  davyraitt/mcp_bedrock_media_generator:latest

Using Docker Compose

version: "3.8"
services:
  mcp-media-generator:
    image: davyraitt/mcp_bedrock_media_generator:latest
    container_name: mcp-media-generator
    restart: unless-stopped
    ports:
      - "8961:8961"
    environment:
      - AWS_ACCESS_KEY_ID=your_access_key
      - AWS_SECRET_ACCESS_KEY=your_secret_key
      - AWS_REGION=us-east-1

LibreChat Integration

Add to your LibreChat librechat.yaml:

mcpServers:
  media-creator:
    type: sse
    url: http://localhost:8961/sse
    timeout: 120000
    initTimeout: 30000
    serverInstructions: |
      Use these tools for AI media generation with Amazon Nova models:
      - Generate high-quality images using Amazon Nova Canvas
      - Create videos using Amazon Nova Reel
      - Customize dimensions, quality, and creative parameters
      - Images are displayed directly in the chat interface
      - Support for various aspect ratios and creative styles

Environment Variables

Variable Description Required Default
AWS_ACCESS_KEY_ID AWS Access Key ID Yes -
AWS_SECRET_ACCESS_KEY AWS Secret Access Key Yes -
AWS_REGION AWS Region No us-east-1

Available Tools

create-image

Generate images using Amazon Nova Canvas

Parameters:

  • prompt (required): Text description of the image
  • negative_prompt (optional): What to avoid in the image
  • quality (optional): "standard" or "premium"
  • width (optional): Image width (320-4096)
  • height (optional): Image height (320-4096)
  • seed_value (optional): Seed for reproducible results

create-video

Generate videos using Amazon Nova Reel

Parameters:

  • prompt (required): Text description of the video

Prerequisites

  • AWS Account with Bedrock access
  • Nova Canvas and Nova Reel models enabled in your AWS region
  • Appropriate IAM permissions for Bedrock access

Development

Building from Source

# Clone the repository
git clone <your-repo-url>
cd mcp_bedrock_media_generator

# Build Docker image
docker build -t davyraitt/mcp_bedrock_media_generator:latest .

# Run locally
docker run -p 8961:8961 \
  -e AWS_ACCESS_KEY_ID=your_key \
  -e AWS_SECRET_ACCESS_KEY=your_secret \
  davyraitt/mcp_bedrock_media_generator:latest

Auto-Deploy Script

Use the included deploy.sh script to automatically build and push updates:

chmod +x deploy.sh
./deploy.sh

Troubleshooting

Common Issues

  • "Model not found" errors: Ensure Nova Canvas/Reel models are enabled in your AWS region
  • Permission denied: Check IAM permissions for Bedrock access
  • Connection timeout: Increase timeout values in LibreChat configuration
  • Images not displaying: Verify you're using the latest version with MCP image format support

Health Check

The server provides a health check endpoint:

GET http://localhost:8961/sse

License

MIT License - see LICENSE file for details

Contributing

  1. Fork the repository
  2. Create a feature branch
  3. Make your changes
  4. Run tests
  5. Submit a pull request

Support

For issues and support:

  • GitHub Issues: Create an issue
  • Docker Hub: davyraitt/mcp_bedrock_media_generator

About

Fixed version of an MCP server for generating images with MCP / Bedrock

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages