Back to search
McpWorkshop-VSLive-Orlando-2026
MCP serverGitHub
End-to-End MCP Server Development in .NET: workshop repository for VS Live Orlando 2026
Safety scan
A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.
Checked today
Badge for your README
Markdown
[](/listings/loekd-consultancy--mcpworkshop-vslive-orlando-2026)HTML
<a href="/listings/loekd-consultancy--mcpworkshop-vslive-orlando-2026"><img src="/api/listings/loekd-consultancy--mcpworkshop-vslive-orlando-2026/badge.svg" alt="findsafeskills safety rating" /></a>Install
i
What was scanned
The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.
README.md · README · 74 lines
- # End-to-End MCP Server Development in .NET — From Local Debugging to Azure
- A hands-on workshop for **VS Live**, November 20, 2026 (8 hours, intermediate–advanced).
- By **Loek Duys** (Microsoft MVP, LoekD Consultancy) and **Alex Thissen**.
- You will build a single **MCP server in ASP.NET Core** and walk it across the chasm
- between "works on my laptop" and "survives production" — closing one engineering **gate**
- at a time: make it work, debuggable, observable, resilient, secure, and deployable to Azure.
- Everything runs in **GitHub Codespaces** — bring only a laptop with a modern browser.
- ## Repository layout
- | Path | Contents |
- |------|----------|
- | [`docs/storyline/`](./docs/storyline/) | The workshop narrative — overview + one file per chapter |
- | [`src/`](./src/) | The Travel Booking MCP server (ASP.NET Core, .NET 11) — the finished solution |
- | [`labs/`](./labs/) | Step-by-step hands-on labs, one per gate, plus the scripts that reset the code to the start of each lab |
- | [`infra/attendee/`](./infra/attendee/) | Scripts that provision each attendee's Azure AI Foundry agent |
- | [`.devcontainer/`](./.devcontainer/) | The Codespaces dev container |
- | [`ARCHITECTURE.md`](./ARCHITECTURE.md) | How the code works: component diagram, one tool call end to end, decisions |
- ## What you build
- A **Travel Booking MCP server**: six tools an AI agent calls to search flights, hold a
- seat, and confirm or cancel a booking, backed by PostgreSQL. Seat holds expire after a short
- time, which is what makes the tracing, resilience and idempotency lessons real.
- ```mermaid
- flowchart LR
- agent["AI agent<br/>or MCP Inspector"] -->|"MCP over HTTP"| server["Travel Booking<br/>MCP server"]
- server --> db[("PostgreSQL")]
- agent -.->|"token (Lab 5)"| idp["Identity provider"]
- server -.->|"validates token"| idp
- ```
- `src/` holds the finished server. Each lab starts from an earlier state of that code and
- adds one gate's worth of it back; [`src/README.md`](./src/README.md) describes the app and
- [`ARCHITECTURE.md`](./ARCHITECTURE.md) explains how a request moves through it.
- ## The storyline at a glance
- > **"Crossing the Chasm: From Demo to Production."**
- Start here: **[docs/storyline/00-overview.md](./docs/storyline/00-overview.md)**.
- | # | Chapter | Gate | Lab |
- |---|---------|------|-----|
- | 00 | Welcome & Setup | — | [Lab 0](./labs/lab-00-setup.md) |
- | 01 | Foundations: Agentic AI, LLMs/SLMs & MCP | — | — |
- | 02 | Make It Work | Gate 1 | [Lab 1](./labs/lab-01-make-it-work.md) |
- | 03 | Make It Debuggable | Gate 2 | [Lab 2](./labs/lab-02-make-it-debuggable.md) |
- | 04 | Make It Observable | Gate 3 | [Lab 3](./labs/lab-03-make-it-observable.md) |
- | 05 | Make It Resilient | Gate 4 | [Lab 4](./labs/lab-04-make-it-resilient.md) |
- | 06 | Make It Secure | Gate 5 | [Lab 5](./labs/lab-05-make-it-secure.md) |
- | 07 | Make It Deployable (Azure) | Gate 6 | [Lab 6](./labs/lab-06-make-it-deployable.md) |
- | 08 | Wrap-Up & Production Checklist | — | — |
- ## Key learnings
- Agentic AI and the Model Context Protocol; modern tooling for distributed apps
- (Dev Tunnels, Dev Proxy); the Microsoft MCP libraries for .NET and ASP.NET Core;
- testing MCP servers locally and from the cloud; authentication and authorization;
- hosting and deployment to Microsoft Azure; and OpenTelemetry instrumentation for
- tracing agentic AI solutions.
- ## Status
- The storyline, the demo app and the labs are in place and are being refined through lab
- dry runs; the slide deck is still to come.
- ## License
- [MIT](./LICENSE).
Ratings and comments
Sign in to rate this skill.
Sign inFree to use. If it’s helped, pass it on.