MCP server for AI image generation and editing using Gemini 3 Pro Image Preview. Enables AI assistants to create and modify images. Integrates with Cursor, Codex, and other developer tools. Supports Claude agents.
git clone https://github.com/shinpr/mcp-image.gitmcp-image is an MCP server that transforms simple text descriptions into high-quality images using Google Gemini or OpenAI. It automatically enriches your prompts with photographic and artistic details—lighting, composition, atmosphere—so you don't need prompt engineering skills. The server supports image editing via natural language instructions, multiple quality tiers from fast iteration to maximum fidelity, and outputs up to 4K resolution in flexible aspect ratios. Features include character consistency for storyboards, Google Search grounding for factual accuracy, and multi-image blending for composite scenes. Works with Cursor, Claude Code, Codex, and any MCP-compatible tool.
[{"step":"Install and configure the mcp-image server in your development environment. Ensure it’s connected to your AI assistant (e.g., Cursor, Codex, or Claude).","tip":"Check the mcp-image documentation for setup instructions specific to your IDE. For Cursor, use the MCP sidebar to add the server."},{"step":"Prepare your image request by specifying either a generation or editing task. For generation, describe the image in detail. For editing, provide the source image path and adjustments.","tip":"Use descriptive prompts with style references (e.g., 'in the style of Studio Ghibli') and technical specs (e.g., resolution, aspect ratio) for better results."},{"step":"Send the request to the mcp-image server via your AI assistant. For example, in Cursor, use the MCP command `/mcp-image [your_prompt]` or integrate it into a custom agent workflow.","tip":"Break complex requests into smaller steps if the server supports incremental updates (e.g., generate base image first, then edit)."},{"step":"Review the output and iterate if needed. Use the server’s feedback to refine prompts or adjustments.","tip":"Save intermediate versions of edited images to avoid overwriting originals. Use version control (e.g., Git) for generated assets in collaborative projects."},{"step":"Integrate the final image into your workflow (e.g., save to a project folder, embed in documentation, or use in a design tool).","tip":"For developer tools like Codex, use the image URI returned by mcp-image to reference the file in your code or documentation."}]
Create product mockups and marketing visuals for e-commerce
Generate storyboard sequences with consistent character appearance
Edit and transform existing images with natural language instructions
Produce high-resolution images for print and professional publications
No install command available. Check the GitHub repository for manual installation instructions.
git clone https://github.com/shinpr/mcp-imageCopy the install command above and run it in your terminal.
Launch Claude Code, Cursor, or your preferred AI coding agent.
Use the prompt template or examples below to test the skill.
Adapt the skill to your specific use case and workflow.
Use the mcp-image server to generate or edit an image based on the following request: [DESCRIBE_IMAGE_REQUEST]. If editing, include the path to the source image: [SOURCE_IMAGE_PATH]. Specify any required adjustments like style, resolution, or modifications. For example: 'Generate a high-resolution cyberpunk cityscape in the style of Blade Runner, with neon lights and flying cars. Use mcp-image to create this image.'
```json
{
"image_generation_request": {
"prompt": "A futuristic cyberpunk cityscape at night, towering skyscrapers with glowing neon signs in Japanese and English, dense urban sprawl with flying cars and holographic billboards, rain-soaked streets reflecting the neon lights, cinematic lighting with deep shadows and vibrant colors, ultra-high resolution, 8K quality",
"style": "cyberpunk",
"resolution": "8192x4320",
"model": "gemini-3-pro-image-preview",
"output_format": "png"
},
"image_editing_request": {
"prompt": "Add a vintage 1980s filter with sepia tones, increase contrast by 20%, and overlay a subtle grid pattern in the background",
"source_image_path": "/users/ai/desktop/photo.jpg",
"adjustments": {
"contrast": 20,
"filter": "vintage-sepia",
"overlay": "grid-pattern"
}
}
}
```
**Generated Image Details:**
- **Resolution:** 8192x4320 pixels (8K UHD)
- **Style:** Cyberpunk with neon aesthetics
- **Features:** Towering skyscrapers, flying cars, holographic billboards, rain-soaked streets with reflections
- **Color Palette:** Deep blues, purples, and vibrant neon accents (pink, cyan, yellow)
- **Lighting:** Cinematic with high contrast between shadows and bright neon lights
**Editing Example:**
For the vintage photo edit, the AI applied a sepia filter, boosted contrast to enhance details in the shadows, and added a subtle grid overlay to give it a retro-computer aesthetic. The original photo was a portrait of a 1980s office worker, now transformed into a nostalgic, high-contrast image with a tech-inspired background pattern.
**Output Location:**
- Generated images are saved to `/users/ai/outputs/[timestamp]_cyberpunk_cityscape.png`
- Edited images overwrite the source file unless specified otherwise.Google's multimodal AI model and assistant
The AI Code Editor for productive developers
Get more done every day with Microsoft Teams – powered by AI
Automate your spreadsheet tasks with AI power
Agentic AI Workflow platform
Connected workspace for docs, wikis, and projects
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan