Image Generation MCP Server

A FastMCP server for generating and editing images using OpenAI's gpt-image-1 and Azure FLUX.1-Kontext-pro models. This component is part of the AppKit platform and is designed to be integrated into the main application.

Features

Text-to-Image Generation: Create images from natural language prompts using multiple AI models
Image Editing & Inpainting: Edit existing images with text prompts and optional masks (gpt-image-1)
Multiple Formats: Output as PNG, JPEG, or WEBP with customizable quality
Prompt Enhancement: Auto-refine prompts via LLM for better results

Configuration

This component uses AppKit's configuration system (appkit_commons) and is configured via MCPImageGeneratorConfig. The settings are loaded from configuration/config.yaml or environment variables mapped by appkit_commons.

Settings

Setting	Description	Default
`backend_server`	URL of the backend server (for retrieving images)	`http://localhost:8000`
`max_file_size_mb`	Maximum allowed file size for input images	`10`
`max_images_to_keep`	Storage retention limit for generated images	`50`
`generator`	Active image generator backend (`azure` or `google`)	`azure`
`azure_api_key`	Azure OpenAI API key	`None`
`azure_base_url`	Azure OpenAI endpoint URL	`None`
`azure_prompt_optimizer`	LLM model used for prompt enhancement (Azure)	`gpt-5-mini`
`azure_image_model`	Image generation model identifier (Azure)	`FLUX.1-Kontext-pro`
`google_api_key`	Google AI API key	`None`
`google_prompt_optimizer`	LLM model used for prompt enhancement (Google)	`gemini-2.0-flash-001`
`google_image_model`	Image generation model identifier (Google)	`imagen-4.0-generate-preview-06-06`
`auth_tokens`	List of MCP tokens and scopes for authentication	`[]`

Integration

This module is designed to be integrated into the main AppKit application rather than running standalone. The create_image_mcp_server function returns a configured FastMCP instance which is mounted by the main application.

Usage in AppKit

In app/app.py, the server is initialized effectively as follows:

from appkit_commons.registry import service_registry
from appkit_user.authentication.services import get_verifier
from appkit_mcp_image.server import create_image_mcp_server, init_generators
from appkit_mcp_image.configuration import MCPImageGeneratorConfig

# ... inside initialization ...
image_mcp_config = service_registry().get(MCPImageGeneratorConfig)
_generators = init_generators(image_mcp_config)

# Create and mount the MCP server
servers["/image"] = create_image_mcp_server(
    _generators[image_mcp_config.generator],
    auth=get_verifier(),
)

Tools & API

generate_image

Create images from text descriptions.

Parameters:

Parameter	Type	Default	Description
`prompt`	string	required	Image description (max 32,000 chars)
`size`	string	`1024x1024`	Dimensions: `1024x1024`, `1536x1024`, `1024x1536`, or `auto`
`output_format`	string	`jpeg`	Output format: `png`, `jpeg`, or `webp`
`seed`	integer	`0`	Random seed for reproducibility (0 = random)
`enhance_prompt`	boolean	`true`	Auto-enhance prompt via LLM
`background`	string	`auto`	Background: `transparent`, `opaque`, or `auto`

Example:

generate_image(
    prompt="A serene mountain landscape at sunset with golden light reflecting off a lake",
    size="1536x1024",
    output_format="png",
    enhance_prompt=True
)

edit_image

Edit existing images with text prompts and optional masks for inpainting.

Parameters:

Parameter	Type	Default	Description
`prompt`	string	required	Description of desired edits (max 32,000 chars)
`image_paths`	array	required	Image URLs, file paths, or base64 data URLs (max 16)
`mask_path`	string	optional	Optional mask image for inpainting (transparent areas indicate edit zones)
`size`	string	`auto`	Output dimensions
`output_format`	string	`jpeg`	Output format: `png`, `jpeg`, or `webp`
`background`	string	`auto`	Background setting

Example:

edit_image(
    prompt="Add a vibrant rainbow across the sky",
    image_paths=["https://example.com/landscape.jpg"],
    mask_path="https://example.com/sky_mask.png",
    output_format="png"
)

Image Input Formats

Supported input methods for image_paths parameter:

HTTP/HTTPS URLs: https://example.com/image.jpg
Local file paths: /path/to/image.png
Base64 data URLs: data:image/png;base64,iVBORw0KG...

Inpainting with Masks

For precise control over edits, use mask images:

Create a PNG image with alpha transparency
Transparent areas (alpha=0) mark regions to edit
Opaque areas remain unchanged
Mask dimensions must match the input image

License

This project is licensed under the MIT License - see LICENSE.md for details.

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

appkit_mcp_image-1.13.7.tar.gz (20.6 kB view details)

Uploaded Aug 25, 2026 Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

The dropdown lists show the available interpreters, ABIs, and platforms. Enable javascript to be able to filter the list of wheel files.

appkit_mcp_image-1.13.7-py3-none-any.whl (16.4 kB view details)

Uploaded Aug 25, 2026 Python 3

File details

Details for the file appkit_mcp_image-1.13.7.tar.gz.

File metadata

Download URL: appkit_mcp_image-1.13.7.tar.gz
Upload date: Aug 25, 2026
Size: 20.6 kB
Tags: Source
Uploaded using Trusted Publishing? No
Uploaded via: uv/0.12.5 {"installer":{"name":"uv","version":"0.12.5","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

File hashes

Hashes for appkit_mcp_image-1.13.7.tar.gz
Algorithm	Hash digest
SHA256	`9ee4c5c09d210f109a4b9138c7e275f86884b2b55e0a197af55fa1a983da916f`
MD5	`e13491bdca37869fcd9bd61235e79b2e`
BLAKE2b-256	`6ade80d37178f7ab86165740a2ca3e1c099b302b4ff8e74d22d812fb5a6ace18`

See more details on using hashes here.

File details

Details for the file appkit_mcp_image-1.13.7-py3-none-any.whl.

File metadata

Download URL: appkit_mcp_image-1.13.7-py3-none-any.whl
Upload date: Aug 25, 2026
Size: 16.4 kB
Tags: Python 3
Uploaded using Trusted Publishing? No
Uploaded via: uv/0.12.5 {"installer":{"name":"uv","version":"0.12.5","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

File hashes

Hashes for appkit_mcp_image-1.13.7-py3-none-any.whl
Algorithm	Hash digest
SHA256	`4d43e03c7f03d1b9c8337373506e49f7d38da9d5464bc62065b581b8cacf5746`
MD5	`c9e59ac993ca628a0c964dc88edc4b46`
BLAKE2b-256	`c4ff1e73668708c67c9e2ce282810fdb778db89c14f12d5383d165b9394789ce`