From Zero to Production MCP Server: workshop repo (MCP Dev Summit Toronto 2026)
View on GitHub ↗Zero to Production MCP: turn any API into an agent tool
Workshop companion for "From Zero to Production MCP Server", MCP Dev Summit Toronto, 5 Oct 2026. A fictional Toronto used-car dealer is the showcase; the patterns are the point. All data is synthetic.
Build Secure Ship --- --- --- An MCP server in front of an internal REST API, one stage at a time Entra sign-in, a permission per tool, human confirmation, secrets the model never sees Container Apps in a private network behind API Management, CI/CD with GitHub Actions
Contents: Follow the workshop · Workshop materials · Why MCP · How MCP talks · One request, end to end · The build · Who can do what · Every call passes these gates · Wrong tool · Human confirmation · When the backend fails · Ship it · See every call · From one server to hundreds · Run it · Deep dive
Everything in the session runs on your laptop: a mock dealer system, the MCP server you build, local test tokens and MCP Inspector. No Azure account needed. The Azure part at the end runs on the presenter's environment; deploy your own later with workshop/deploy-azure.ipynb (Deploy to Azure).
Before the session (downloads are slow on shared conference Wi-Fi): install git, Python 3.11+, uv and Node.js (for MCP Inspector), then run:
On Windows, use WSL or Git Bash for the scripts.
During the session, open three terminals in the repo:
Terminal Command What it runs --- --- --- 1 ./scripts/demo.sh dms The mock dealer system on :8081 2 ./scripts/demo.sh live The MCP server you build (src/live/server.py) on :8080, reloads on save 3 ./scripts/demo.sh inspector
From the project README.
Add the radar badge to your README — it shows your project was picked up by MCP Radar and links to this page:
[](https://mcp.liqiwa.com/s/hkaanturgut--zero-to-production-mcp.html)
Stop your AI from making things up — it proposes, deterministic tools decide, every claim checked against ground truth with evidence. Grounded facts and context survive resets. Reverse engineering is
zhuyansen/new-term-radarDaily radar for new AI terms: names of new models, products and repos, caught on X, Hugging Face, GitHub and sitemaps before keyword tools list them.
yzg1226/mcd-party-officer🍟 麦当劳门店派对成团调度 Skill — 把派对从「策划」推进到「办得成」：成团概率雷达 / 包场vs拼团 / 择日黄历 / 儿童营养指数 / 招募卡与邀请函
fellowgeek/mcp-memoryAn OKF-backed Model Context Protocol (MCP) server delivering persistent long-term memory and SQLite FTS5 search for AI agents.
hirotomasato/yowesGenerate realistic teacher documents (ID cards, licenses, letters) for 13 countries via MCP.
hujjun/mcd-saver-strategist麦当劳省钱点餐助手（麦门精算师）｜麦当劳优惠券组合优化 Skill：拉取麦当劳菜单与全部可用优惠券，算出最省的点餐组合，巨无霸套餐 ¥47 → ¥21.9（省 53%）。McDonalds MCP（麦当劳 MCP）coupon optimizer · McDonalds China MCP Server · Agent Skill for WorkBuddy · 麦当劳程序员创意开发大赛参赛作品
The top new MCP servers of the week, every Monday. No spam, unsubscribe anytime.