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.
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:
Configure Your AI Assistant
Add Pixault to your MCP client configuration. For Claude Desktop, add to claude_desktop_config.json:
For Cursor or other editors, follow their MCP server configuration documentation using the same command and environment variables.
Environment Variables
Variable | Required | Description |
|---|---|---|
| No | API endpoint (default: |
| Yes | Your project identifier |
| Yes | API key for authentication |
| No | HMAC secret for signed URL generation |
Available Tools
The MCP server provides 14 tools across 5 categories:
Image Management
Tool | Description |
|---|---|
| Upload an image or video from a local file path |
| List images with cursor-based pagination |
| Delete an image and all cached variants |
| Generate a CDN URL with transformation parameters |
Metadata
Tool | Description |
|---|---|
| Get dimensions, size, Schema.org fields, and tags |
| Update name, description, keywords, author, copyright, license |
| Get the JSON-LD structured data endpoint URL |
Named Transforms
Tool | Description |
|---|---|
| List all named transform presets |
| Get details of a specific preset |
| Create or update a named transform |
| Delete a named transform |
Plugin Marketplace
Tool | Description |
|---|---|
| List all available marketplace plugins |
| List plugins with activation status for your project |
| Activate a plugin for your project |
| Deactivate a plugin |
Documentation
Tool | Description |
|---|---|
| Search Pixault documentation by topic or keyword |
| List all available documentation topics |
Usage Examples
Once configured, you can use natural language with your AI assistant:
Image management:
Metadata:
Transforms:
Plugins:
Documentation:
Supported Formats
The UploadImage tool supports:
Images: JPEG, PNG, WebP, GIF, AVIF, SVG
Video: MP4, WebM, MOV (auto-thumbnail generation)