mcpField
P
PatchDesign.AI

PatchDesign.AI

Remote
patchdesign.aiMedia & Design0Updated 1w ago

About PatchDesign.AI

PatchDesign.AI is an MCP server for creating custom patch designs from text descriptions or image URLs. It also supports revisions, price quotes, and order links, without placing or charging for an order in chat.

Key features

  1. 01Generate designs from text or image URLs
  2. 02Edit designs using plain language
  3. 03Quote prices by type, size, and quantity
  4. 04Create links with order details
  5. 05Save designs by connecting an account

Use cases

  1. 01Create a custom patch for a team or event
  2. 02Revise a design through chat
  3. 03Price a production run before ordering
  4. 04Open a design to review and order

MCP support

Exposes patch design, editing, pricing, account linking, and order-link tools through a remote Streamable HTTP MCP server. Clients can connect using its server URL without credentials to start.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://patchdesign.ai/api/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Tools

5
ParameterTypeRequiredDescription
widthInnumberNoWidth in inches, 1 to 12.5. Default 3.
heightInnumberNoHeight in inches, 1 to 12.5. Default 3.
quantityintegerNoHow many patches. Default 100. Use 1 for a single patch.
patchTypeenumNoPatch type: embroidered (default), woven, chenille, printed, pvc-2d or pvc-3d. Iron-on, velcro and sew-on are backings on an embroidered patch, so use embroidered for those.embroidered | woven | chenille | printed | pvc-2d | pvc-3d

Details

From the project README

PatchDesign.AI MCP server

PatchDesign.AI designs custom patches in your AI chat. Describe a patch for your team, troop, club, department, event or brand, and the server generates a production-ready design and shows it in the chat. Ask for changes in plain words, ask what a run would cost, and get a link that opens your design on patchdesign.ai to order it.

This repository holds the server's documentation. The server itself runs on patchdesign.ai; its source lives in PatchDesign.AI's private codebase.

What the server does

ToolWhat it does
quote_patchPrices embroidered, woven, chenille, printed and PVC patches from the live production price table, by size and quantity.
design_patchGenerates a patch design from a description, or from the URL of a logo, sketch or photo. Takes about 15 to 30 seconds.
edit_patchChanges a design in plain words ("make the border gold"). Earlier versions stay.
order_linkA link that opens the design on patchdesign.ai with the type, size and quantity set. Nothing is ordered or charged from the chat.
sign_in_linkConnects a PatchDesign.AI account, which saves your designs to it.

In clients that draw MCP Apps (GitHub Copilot in VS Code, ChatGPT, Claude, Cursor) a design shows as a card: the preview, its design reference (D-...), type and size, and an order button. Other clients show the preview image and the links, or the result's text and links.

Previews carry a light PatchDesign.AI mark; the clean production file comes with the order. Each chat can make up to ten designs before you connect an account.

Configure

The server needs no credentials. Add it by URL wherever your client takes a remote MCP server; https://patchdesign.ai/connect has the menu path for each one, checked against the vendors' help pages.

GitHub Copilot in VS Code: add the server to .vscode/mcp.json in a workspace (or to your user mcp.json through MCP: Open User Configuration):

{
  "servers": {
    "patchdesign": {
      "type": "http",
      "url": "https://patchdesign.ai/api/mcp"
    }
  }
}

Then start it from the MCP: List Servers command, or from the server's CodeLens, and check that Copilot's Agent mode shows five tools.

Claude Code:

claude mcp add --transport http patchdesign https://patchdesign.ai/api/mcp

Cursor: add the server to ~/.cursor/mcp.json (every project) or .cursor/mcp.json (one project):

{
  "mcpServers": {
    "patchdesign": {
      "url": "https://patchdesign.ai/api/mcp"
    }
  }
}

ChatGPT, Claude, Gemini, Grok, Perplexity, Le Chat: add a custom connector with the server URL; the steps for each are on https://patchdesign.ai/connect.

Use

Ask in plain words. Things to try:

  • Design a 3 inch embroidered patch for Troop 42 with an eagle in navy and gold.
  • How much are 100 embroidered 3 inch patches?
  • Make the border gold and change the text to EST. 2026.
  • I want to order 50 of this one.

A design takes about 15 to 30 seconds. Each version gets its own preview and D- number. To change a design in a later turn, refer to it by its D- number in the same chat. When you want the patches made, the order link opens the design on patchdesign.ai, where you review the order and pay. Nothing is ordered or charged from the chat, and no delivery date is promised there: production takes about 10 business days after artwork approval, plus shipping.

After ten designs in a chat the server offers a sign-in link. Connecting a PatchDesign.AI account keeps the designs coming and saves them to the account at https://patchdesign.ai/my-designs.

Privacy and terms

Designs made through the chat are kept with the conversation that made them, so a later turn can change or order them. PatchDesign.AI's privacy policy (Section 3A covers the AI-chat connector) and terms apply.

Support

[email protected], or https://patchdesign.ai/contact.

License

The documentation in this repository is under the MIT License (see LICENSE). The PatchDesign.AI name and logo belong to Kinor LLC.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.