Printify MCP
Enables AI assistants to run a print-on-demand store by creating products on Printify, pricing from production cost, publishing to Shopify and its sales channels, and generating ad creative from mockups.
README
Printify MCP
An MCP server for running a print-on-demand store from an AI assistant: create products on Printify, price them from their production cost, publish them, push them to every Shopify sales channel, and render ad creative from the product mockups.
Fork of TSavo/printify-mcp, substantially extended. See Differences from upstream.
Why it exists
The Printify API alone gets you about two thirds of a working pipeline. The gaps that this fills, each found by hitting it in production:
- Prices are hand-typed. Printify reports each variant's production cost, but nothing
derives a price from it.
set_pricesdoes, in margin, markup or fixed mode, rounding upward so rounding never eats the target margin. - Publishing reaches one channel. Printify publishes to Shopify's Online Store
publication and stops. TikTok, Google & YouTube, Pinterest, Shop and POS stay
unpublished until something else pushes them.
publish_to_sales_channelsdoes that. - Printify guesses the product type wrong. A flask arrives typed "Mug", a candle as
"Home Decor". Downstream channels map their taxonomy from that field.
set_product_categoryderives it from Shopify's own category. - Ad creative is manual.
create_product_carouselandcreate_product_adsbuild vertical posts from the product's real mockups.
Install
git clone https://github.com/<you>/printify-mcp.git
cd printify-mcp
npm install
npm run build
cp .env.example .env # then fill in PRINTIFY_API_KEY
Register it with your MCP client. For Claude Code:
{
"mcpServers": {
"printify": {
"type": "stdio",
"command": "node",
"args": ["/absolute/path/to/printify-mcp/dist/index.js"]
}
}
}
The server resolves .env from its own package directory, not the working directory, so
it works no matter where the host launches it from.
Only PRINTIFY_API_KEY is required. Shopify and Replicate credentials unlock the
sales-channel and image-generation tools respectively; without them those tools refuse
with setup instructions rather than failing obscurely.
Tools
Shops and catalog — get_printify_status, list_shops, switch_shop,
get_blueprints, get_blueprint, get_print_providers, get_variants,
get_defaults, set_default
Products — list_products, get_product, create_product, update_product,
delete_product, publish_product, unpublish_product, get_product_mockups,
upload_image
Pipeline — create_and_publish runs create → price → publish → fan out to sales
channels → fix category in one call, reporting each stage separately so a partial failure
tells you exactly how far it got.
Pricing — set_prices (margin / markup / fixed, dryRun to preview)
Sales channels (Shopify credentials) — list_sales_channels,
publish_to_sales_channels, get_channel_sales, list_channel_products
Categories (Shopify credentials) — suggest_product_category,
set_product_category, audit_product_categories
Creative — create_product_carousel (numbered 1080×1920 photo slides),
create_product_ads (three vertical video variants; needs ffmpeg on PATH)
Orders — get_orders, get_order, get_orders_by_day, get_orders_by_hour,
get_orders_by_month, get_orders_in_range
Webhooks — list_webhooks, create_webhook, delete_webhook
Images (Replicate) — generate_image, generate_and_upload_image
Docs — how_to_use with a topic (pipeline, product_creation, blueprints,
print_providers, variants, images, publishing, image_generation) returns the
reference an agent should read before guessing.
Notes worth knowing
- Never publish a product twice. Publishing an already-published product creates a
second channel listing and orphans the first — no Printify link, still purchasable,
unfulfillable.
publish_productrefuses unlessrepublishis passed. external.idis not reliably a Shopify product ID. Immediately after publishing it can carry aprintify.mehandle. The channel tools wait until it points at the configured store; passshopifyProductIdto skip the lookup.- Errors are real errors. No tool returns empty or placeholder data to paper over a failed request. A bad API key fails loudly at startup rather than quietly serving fabricated shops.
- Creative is built from product photography and text — no generative model touches it, so the output carries no AI provenance metadata.
Differences from upstream
Added: order history and range queries, cost-based pricing, Shopify sales-channel
publishing, channel sales reporting, product-category correction, photo carousels and
video ads, tags/key features/sales-channel properties on create, print placement
(x/y/scale/angle plus image stacking), duplicate-title guard, and the create_and_publish
pipeline.
Fixed: list_products passed (limit, page) into a (page, limit) signature; the
products list silently returned an empty page on failure; a rejected API key produced two
fake shops and reported success; update_product forwarded camelCase variant keys the API
ignores, so price updates reported success and changed nothing; publish_product sent a
partial payload where Printify requires all seven flags.
Requirements
Node 18+. ffmpeg on PATH for create_product_ads (the carousel tool does not need it).
License
ISC, as upstream. See LICENSE — copyright TSavo, retained for the original work.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.