madeforai-supplychain-skill
LocalAbout madeforai-supplychain-skill
madeforai-supplychain-skill is an open-source connector that forwards authorized manufacturing requests from MCP clients to MadeForAI's hosted intake service. Requests enter a manual-review queue; the connector does not provide production control or payment authority.
Key features
- 01Forward authorized manufacturing requests
- 02Generate artwork briefs and quote requests
- 03Create and confirm order drafts
- 04Start a prototype pipeline
- 05Retrieve task status
Use cases
- 01Submit a manufacturing task for human review
- 02Prepare artwork and quotation requests
- 03Draft an order before confirmation
MCP support
Exposes nine public tools through MCP and forwards requests to `https://api.madeforai.net/mcp`. MCP clients can use the hosted endpoint directly or run the stdio connector with an access token.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
npx -y madeforai-supplychain-connectorMADEFORAI_ACCESS_TOKENSecretMadeForAI client access token. Request a token from MadeForAI and keep it secret.Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEMadeForAI Hosted Supply Chain Connector
This open-source MCP connector forwards authorized manufacturing task requests from Codex, Claude, Gemini, Cursor, and other MCP clients to the MadeForAI hosted order-intake service.
Trust boundary
The connector contains only stdio transport, the nine public tool names, and a fixed upstream endpoint: https://api.madeforai.net/mcp.
It does not contain the order database, Chinese production workspace, production credentials, supplier records, payment-confirmation authority, shipment authority, or fulfillment administration. Forks remain usable under MIT, but they do not gain access to MadeForAI-hosted tasks or private production systems.
Run with npx
Set your MadeForAI client access token, then run the connector directly from npm:
npx --yes [email protected]
For MCP clients, use npx as the command:
{
"mcpServers": {
"madeforai": {
"command": "npx",
"args": ["--yes", "[email protected]"],
"env": {
"MADEFORAI_ACCESS_TOKEN": "replace-with-your-32-character-or-longer-token"
}
}
}
}
The command exits with a clear error until MADEFORAI_ACCESS_TOKEN is set. Tokens must contain at least 32 characters.
Manual installation
As a fallback, download madeforai-supplychain-connector-0.5.2.tgz from the matching GitHub Release, then install it:
npm install --global ./madeforai-supplychain-connector-0.5.2.tgz
madeforai-supplychain
Build from source
git clone https://github.com/amplichina/madeforai-supplychain-skill.git
cd madeforai-supplychain-skill
npm ci
npm run build
npm test
MCP client configuration
Request a client access token from MadeForAI. Keep it out of source code, screenshots, issue reports, and public examples.
For an npx-based installation:
{
"mcpServers": {
"madeforai": {
"command": "npx",
"args": ["--yes", "[email protected]"],
"env": {
"MADEFORAI_ACCESS_TOKEN": "replace-with-your-32-character-or-longer-token"
}
}
}
}
For a source checkout, use node with the absolute path to dist/src/index.js.
Clients with native remote MCP support may connect directly:
{
"url": "https://api.madeforai.net/mcp",
"headers": {
"Authorization": "Bearer replace-with-your-client-token"
}
}
Public MCP tools
create_supplychain_taskgenerate_artwork_briefgenerate_quote_requestgenerate_order_draftconfirm_order_draftconfirm_production_feedbackcreate_payment_linkquick_start_prototype_pipelineget_task
Production feedback, payment confirmation, manufacturing progress, quality checks, shipment, and completion remain unavailable through this public connector.
Safety boundaries
MadeForAI does not generate images, run LLM inference, process real payments, or automatically place supplier orders in this release. A submitted task enters a manual-review queue; submission does not mean that review has started, and no response time is guaranteed. Prices and lead times remain unconfirmed until manual feedback is recorded.
License
The connector is MIT licensed. The hosted control plane, production workspace, credentials, databases, and production operations are separate proprietary services.
Comments
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.