OHMS
Exposes Shopify order and inventory management tools via MCP, allowing agents to fetch, update, and print orders without exposing raw Shopify credentials.
README
OHMS - Order Hub Management System
Flauraly Flowers and Plants - Python FastMCP server hosted on Replit Reserved VM.
Purpose
OHMS exposes a small, hardened set of order- and inventory-related tools to Violet (and any other authorized MCP client) over the Model Context Protocol. It centralizes Shopify Admin REST access behind a Bearer-auth gate so agents never touch raw Shopify credentials.
Architecture
+-----------------+ Bearer +------------------------------+
| MCP Client | ---------------> | OHMS (Replit Reserved VM) |
| (Violet, etc.) | /mcp or /sse | |
+-----------------+ | Starlette parent app |
| +-- /health (open) |
| +-- /mcp (Streamable HTTP)|
| +-- /sse (SSE fallback) |
| +-- BearerAuthMiddleware |
+--------------+---------------+
|
v
+------------------------------+
| Shopify Admin REST API |
+------------------------------+
Both /mcp (Streamable HTTP) and /sse (Server-Sent Events fallback) are
mounted simultaneously so any MCP client transport profile works.
Environment Variables
OHMS authenticates to Shopify via the OAuth 2.0 client_credentials grant.
The server never holds a long-lived SHOPIFY_ACCESS_TOKEN; instead it holds a
SHOPIFY_CLIENT_ID + SHOPIFY_CLIENT_SECRET pair and mints a short-lived
access token on demand against https://{shop}.myshopify.com/admin/oauth/access_token.
The minted token is cached in-memory with a 5-minute clock-skew buffer and
re-minted automatically (or on a 401/403 from any subsequent call). All values
are read via os.environ.get(...). Nothing is hardcoded.
| Var | Purpose |
|---|---|
PORT |
TCP port to bind (Replit injects this; defaults to 8080). |
OHMS_API_TOKEN |
Static bearer token required on every non-/health request (client to OHMS auth - separate from Shopify). |
SHOPIFY_STORE_URL |
Shop domain, e.g. flauraly.myshopify.com. |
SHOPIFY_CLIENT_ID |
Shopify app client ID (used for OAuth client_credentials grant). |
SHOPIFY_CLIENT_SECRET |
Shopify app client secret (used for OAuth client_credentials grant). Rotate per Secrets_Registry.md schedule. |
SHOPIFY_API_VERSION |
Pinned Shopify API version, e.g. 2025-01. |
PRINTER_IP |
Local network IP of the receipt printer (stub uses this). |
See .env.example for the placeholder template.
Local Dev (Windows)
OHMS reads .env only when running locally (via python-dotenv). Bootstrap
your .env from the Windows DPAPI-protected secrets store rather than typing
secrets in plaintext:
# 1. Pull each secret from DPAPI into the local .env (PowerShell pseudocode)
$secrets = @("OHMS_API_TOKEN","SHOPIFY_STORE_URL","SHOPIFY_CLIENT_ID","SHOPIFY_CLIENT_SECRET","SHOPIFY_API_VERSION","PRINTER_IP")
foreach ($k in $secrets) {
$v = Unprotect-DpapiSecret -Name $k # your local helper
Add-Content .env "$k=$v"
}
# 2. Run the server
python -m venv .venv
.venv\Scripts\Activate.ps1
pip install -r requirements.txt
python main.py
The server listens on http://0.0.0.0:8080 by default. Probe with:
curl http://localhost:8080/health
# => OHMS OK
Replit Deploy
- Create a Replit project and import this folder.
- In the Secrets pane, set every variable from
.env.example(using the real values - never paste them into any committed file). - Confirm
.replitshowsdeploymentTarget = "reserved_vm"and port8080 -> 80. - Deploy. The public URL is
https://ohms-server.crashzero9.replit.app. - Verify both transports:
curl https://ohms-server.crashzero9.replit.app/health curl -H "Authorization: Bearer $OHMS_API_TOKEN" \ -H "Content-Type: application/json" \ -d '{"jsonrpc":"2.0","method":"tools/list","id":1}' \ https://ohms-server.crashzero9.replit.app/mcp
Tool Registry
| Tool | Source | Notes |
|---|---|---|
get_order(order_id) |
Shopify Admin REST | GET /orders/{id}.json |
list_pending_orders() |
Shopify Admin REST | GET /orders.json?status=open&limit=50 |
update_order_status(order_id, status) |
Shopify Admin REST | PUT /orders/{id}.json (sets tags) |
get_inventory_snapshot() |
Shopify Admin REST | GET /inventory_levels.json?limit=50 |
get_doordash_orders_via_browser() |
Stub | Returns routing dict for browser-automation handoff. |
print_order_ticket(order_id) |
Stub | Reads PRINTER_IP; returns queued status. Driver pending. |
Phase Status
- Phase 1 (current): MVP scaffold, Bearer auth, six tools (4 live + 2 stubs), Replit Reserved VM deploy, basic pytest suite.
- Phase 2 (planned): OAuth 2.1 replacing static bearer, full DoorDash driver, real network printer driver, structured logging with secret redaction, rate limiting.
Security Notes
- No secret values appear in any committed file.
Authorizationheaders and full Shopify response bodies are never logged./healthreturns only the literal stringOHMS OK- no version, env, or path info.- All
httpxcalls have an explicit 30-second timeout.
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.
E2B
Using MCP to run code via e2b.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
Neon Database
MCP server for interacting with Neon Management API and databases
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.