#MCP Server Integration

Pixault includes a Model Context Protocol (MCP) server that makes all image management capabilities accessible to AI assistants like Claude, Cursor, Windsurf, and any other MCP-compatible client.

Source on GitHub

#What is MCP?

The Model Context Protocol is an open standard for connecting AI assistants to external tools and data sources. Pixault's MCP server exposes image upload, transformation, metadata management, and plugin control as tools that AI assistants can call directly.

#Setup

#Install

The MCP server is distributed as a .NET tool:

dotnet tool install --global Pixault.Mcp

#Configure Your AI Assistant

Add Pixault to your MCP client configuration. For Claude Desktop, add to claude_desktop_config.json:

{
  "mcpServers": {
    "pixault": {
      "command": "pixault-mcp",
      "env": {
        "PIXAULT_BASE_URL": "https://img.pixault.io",
        "PIXAULT_PROJECT": "myapp",
        "PIXAULT_API_KEY": "pk_your_api_key",
        "PIXAULT_HMAC_SECRET": "your_hmac_secret"
      }
    }
  }
}

For Cursor or other editors, follow their MCP server configuration documentation using the same command and environment variables.

#Environment Variables

Variable	Required	Description
`PIXAULT_BASE_URL`	No	API endpoint (default: `https://img.pixault.io`)
`PIXAULT_PROJECT`	Yes	Your project identifier
`PIXAULT_API_KEY`	Yes	API key for authentication
`PIXAULT_HMAC_SECRET`	No	HMAC secret for signed URL generation

#Available Tools

The MCP server provides 36 tools across 8 categories:

#Image Management

Tool	Description
`UploadImage`	Upload an image or video from a local file path
`ListImages`	List and search images with cursor-based pagination and filters (category, keyword, author, folder, video)
`FindImageByFilename`	Find an image by its exact original filename and return its ID plus CDN URLs
`ResolveImageUrls`	Resolve many original filenames to CDN URLs in one call
`BuildImageUrl`	Generate a CDN URL with transformation parameters
`GenerateArchive`	Download a zip of one or more images' original files to a local path
`DeleteImage`	Delete an image and all its cached variants
`DeleteImages`	Bulk-delete images by IDs, filenames, folder, search, or category (preview-first)

#Metadata & Organization

Tool	Description
`GetImageMetadata`	Get dimensions, size, Schema.org fields, and tags
`UpdateImageMetadata`	Update name, description, keywords, author, copyright, license, and more
`MoveImage`	Re-file an image into a folder (or the root) — metadata-only, URL unchanged
`RenameImage`	Change an image's display name (Schema.org name) — does not change the publicId or URL
`TagImage`	Add keyword tags (appends by default, or replace)
`ListTags`	List all distinct keyword tags in a project, with per-tag image counts
`StripExif`	Strip EXIF (camera, GPS, timestamps) from an image's stored original
`GetJsonLd`	Fetch the Schema.org JSON-LD document for an image

#Folders

Tool	Description
`ListFolders`	List the folders defined for a project
`CreateFolder`	Create a folder in a project
`DeleteFolder`	Delete a folder from a project

#Named Transforms

Tool	Description
`ListTransforms`	List all named transform presets
`GetTransform`	Get details of a specific preset
`SaveTransform`	Create or update a named transform
`DeleteTransform`	Delete a named transform

#Watermarks

Tool	Description
`ListWatermarks`	List the watermark images stored for a project
`UploadWatermark`	Upload (or replace) a watermark image from a local file
`DeleteWatermark`	Delete a watermark image

#EPS & Vector

Tool	Description
`SplitEpsDesigns`	Split a multi-design EPS file into individual design assets
`ExtractEpsSvg`	Extract a scalable SVG from an EPS/vector source
`GetDerivedAssets`	List assets derived from a source image (split designs, extracted SVG)
`GetProcessingStatus`	Check the status of an EPS split or vector-extraction job

#Plugin Marketplace

Tool	Description
`ListPlugins`	List all available marketplace plugins
`ListProjectPlugins`	List plugins with activation status for your project
`ActivatePlugin`	Activate a plugin for your project
`DeactivatePlugin`	Deactivate a plugin

#Documentation

Tool	Description
`SearchDocs`	Search Pixault documentation by topic or keyword
`ListDocTopics`	List all available documentation topics

#Usage Examples

Once configured, you can use natural language with your AI assistant:

Image management:

"Upload the product photo at /photos/widget.jpg to my Pixault account"

"List my recent images"

"Generate a 400x400 thumbnail URL for img_01JKABC in WebP format"

Metadata:

"Add copyright info to img_01JKABC — author is Jane Smith, year 2026, CC-BY-4.0 license"

"What metadata does img_01JKABC have?"

Organizing:

"Move img_01JKABC into the products/hero folder"

"Tag img_01JKABC with summer, beach, and promo"

"What tags are already in use across this project?"

"Rename img_01JKABC's display name to 'Autumn Collection Banner'"

"Archive img_01JKABC and img_01JKDEF into ./originals.zip"

Transforms:

"Create a named transform called 'hero' with 1920px width, 80% quality, cover fit"

"Show me all my transform presets"

Plugins:

"What plugins are available in the marketplace?"

"Activate the background removal plugin"

Documentation:

"How do I use signed URLs?"

"Show me the upload API documentation"

#Supported Formats

The UploadImage tool supports:

Images: JPEG, PNG, WebP, GIF, AVIF, SVG
Video: MP4, WebM, MOV (auto-thumbnail generation)