This server provides a suite of image manipulation tools via MCP, with 25+ tools covering core operations, drawing, filtering, transformations, metadata, and advanced analysis. Tools include:
Core Operations: raster_info (metadata), raster_convert (format conversion), raster_resize, raster_crop, raster_rotate, raster_adjust (brightness, contrast, etc.).
Drawing & Graphics: raster_text (overlay text), raster_draw (shapes).
Filtering: raster_filter (blur, sharpen, etc.), raster_enhance (auto-enhance).
Transformations: raster_perspective, raster_morphology, raster_balance, raster_padding, raster_channels.
Metadata & Compression: raster_exif, raster_colorspace, raster_blend, raster_compress.
Advanced Analysis: raster_diff (SSIM), raster_histogram, raster_edge (Canny), raster_qr (decoding), raster_bgremove.
Backend: Pillow (default) or ImageMagick for extended formats and advanced filters.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@mcp_images Resize the image to 800x600"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP server for raster image manipulation — Pillow + OpenCV. 25 tools across draw, filter, transform, metadata, and analysis categories.
Requires pi-mcp-bridge to connect to pi.
# Base install (Pillow + ImageMagick/Wand)
pip install git+https://github.com/timaliev/mcp_images.git
# Or via uv:
uv tool install git+https://github.com/timaliev/mcp_images.gitExtra | Packages | Tools enabled |
| scikit-image, pyzbar, rembg |
|
# With analysis extras:
pip install "git+https://github.com/timaliev/mcp_images.git#egg=mcp-images[analysis]"
# Or via uv:
uv tool install "git+https://github.com/timaliev/mcp_images.git#egg=mcp-images[analysis]"ImageMagick must be installed separately via system package manager:
brew install imagemagick # macOS
apt install imagemagick # Debian/UbuntuRelated MCP server: jgkme/kilo-image-gen-mcp
First install pi-mcp-bridge:
pi add mcp-bridge
# or manually:
# git clone https://github.com/timaliev/pi-mcp-bridge ~/.pi/agent/extensions/pi-mcp-bridgeThen configure mcp-images in ~/.pi/agent/settings.json:
{
"mcpBridge": {
"servers": [
{
"name": "raster",
"command": "mcp-images",
"args": [],
"setupCommands": [
"uv tool install --force --python 3.11 \"mcp-images[analysis] @ git+https://github.com/timaliev/mcp_images.git\""
]
}
]
}
}In ~/.mcp.json:
{
"mcpServers": {
"raster": {
"command": "mcp-images",
"args": []
}
}
}Core tools accept an optional backend parameter ("pillow" or "magick").
Backend | Library | Strengths |
| Pillow + OpenCV | Fast, simple, no extra deps |
| Wand / ImageMagick | HEIC/AVIF/GIF, 40+ resize filters, industry-standard unsharp mask |
# Use ImageMagick for HEIC conversion
raster_convert(path, "heic", backend="magick")
# Mitchell filter resize via ImageMagick
raster_resize(path, width=800, backend="magick")
# Back to Pillow for speed
raster_crop(path, 10, 10, 100, 100, backend="pillow")Tool | Description |
| Image metadata: dimensions, format, mode, DPI, file size |
| Convert format (png, jpeg, webp, tiff, bmp). Use |
| Resize by dimensions, scale, or fit mode |
| Crop to rectangle |
| Rotate by degrees |
| Brightness, contrast, saturation, sharpness, gamma |
Tool | Description |
| Overlay text with font size and color |
| Draw shapes: rect, circle, line, arrow |
Tool | Description |
| Blur, sharpen, denoise, grayscale, invert, threshold. Use |
| Auto-enhance: contrast, color, sharpness, all |
Tool | Description |
| 4-point perspective correction |
| Morphological ops: dilate, erode, open, close |
| Auto balance: equalize, autocontrast, auto white |
| Add border/margin with fill color |
| Channel ops: extract, swap, reorder |
| Compress: quality control, metadata stripping |
Tool | Description |
| Extract EXIF metadata as dict |
| Convert: RGB, HSV, LAB, grayscale |
| Alpha-blend two images |
| Find contours with OpenCV, draw on output |
pip install mcp-images[analysis])Tool | Description |
| SSIM structural similarity between two images |
| Channel histogram (R/G/B/all) |
| Canny edge detection |
| Decode QR codes and barcodes |
| Remove background (returns RGBA) |
git clone https://github.com/timaliev/mcp_images.git
cd mcp_images
uv run pytestUnclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
MCP server for NanoBanana AI image generation and editing
MCP server for Midjourney AI image generation and editing
MCP server for Flux AI image generation
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/timaliev/mcp_images'
If you have feedback or need assistance with the MCP directory API, please join our Discord server