Pictify gives AI agents the ability to generate, manage, and A/B test visual content. No code required — just describe what you want.
31 tools across 6 categories
Image Generation
Create images from HTML/CSS, URL screenshots, or reusable templates. PNG, JPEG, WebP up to 4000x4000px. Perfect for OG images, Twitter cards, banners, and product screenshots.
GIF Creation
Generate animated GIFs from CSS animations or record live web pages. Ideal for product demos, animated banners, and loading animations.
PDF Rendering
Single or multi-page PDFs from templates. Invoices, certificates, reports, shipping labels, and ID cards with preset page sizes.
Template System
Create, manage, and render reusable templates with variable placeholders. 50+ expression functions for conditionals, string manipulation, and date formatting.
Batch Operations
Generate up to 100 personalized images in one request. Team badges, event passes, product catalogs, localized marketing assets.
A/B Testing
Built-in experiments with traffic splitting, Thompson Sampling auto-optimization, and event tracking for impressions, clicks, and conversions.
Quick Start
- Get your API key at https://pictify.io/dashboard/api-tokens
- Add the MCP server to your client
- Start generating — just ask your AI assistant to create images, GIFs, or PDFs
Works with Claude (claude.ai, Claude Code, Claude Desktop), Cursor, Windsurf, and any MCP-compatible client.
Note: For Github URL, I'd use the actual GitHub repo (https://github.com/pictify-io/pictify-mcp) instead of the npm link — mcp.so likely expects a GitHub URL for that field.
Server Config
{
"mcpServers": {
"pictify": {
"command": "npx",
"args": [
"-y",
"@pictify/mcp-server"
],
"env": {
"PICTIFY_API_KEY": "your_api_key"
}
}
}
}