agentleFS
MicrosoftKnown publishermicrosoft/sqlcon-eu-2026-demos/.github/skills/run-keynote-demo-1/SKILL.md

run-keynote-demo-1

Skill0 starsChanged 53 days agoMITView on GitHub

Prepare, validate, rehearse, or troubleshoot SQLCon EU 2026 Keynote Demo 1, Caldova Regional Care on Azure Local. Use when running the Demo 1 setup, checking SQL Server 2025 and Foundry Local, validating the Azure Local gateway, starting the transfer-center app, or guiding the locked presenter flow.

The file

Show its rules and text

We split this skill into the rules it gives an agent

14 rules. About 0 are common, 10 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈10

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Never display or read into chat the kubeconfig, Kubernetes API key, database scoped credential secret, database master-key password, or a …

    line 17
  2. Distinctive:

    Never retrieve a Kubernetes Secret directly for inspection.

    line 20
  3. Distinctive:

    Do not provision, restart, scale, or modify Azure Local, AKS Arc, Foundry Local, the model deployment, or portal resources.

    line 22
  4. Very specific:

    Do not bypass TLS validation or add --insecure.

    line 24
  5. Distinctive:

    Ask for explicit confirmation before running an elevated setup, restarting SQL Server, initializing the database, invoking the live model, …

    line 25
  6. Distinctive:

    Never request a secret through chat.

    line 28
  7. Distinctive:

    Check, without printing sensitive contents:

    line 44
  8. Very specific:

    Ask before invoking the live model.

    line 61
  9. Distinctive:

    Run the live AI validation only after separate confirmation:

    line 73
  10. Distinctive:

    Follow the runbook's locked order exactly:

    line 99
  11. Distinctive:

    Start from the failed gate and its exact error.

    line 116
  12. Distinctive:

    Read the owning script before proposing a change.

    line 117
  13. Distinctive:

    Prefer rerunning a documented validation over changing infrastructure.

    line 118
  14. Very specific:

    Do not use Connect-AksArc.ps1 unless direct kubeconfig access failed and the user explicitly asks to troubleshoot Arc Cluster Connect.

    line 119

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Run Keynote Demo 1

Guide the user through the existing Caldova Regional Care demo. Treat the package README and demo runbook as the sources of truth. Use repository scripts as written; do not recreate their logic with ad hoc commands.

Safety Rules

  • Never display or read into chat the kubeconfig, Kubernetes API key, database scoped credential secret, database master-key password, or a literal key from chatLanguageModels.json.
  • Never retrieve a Kubernetes Secret directly for inspection. Only the existing setup and database wrappers may consume it.
  • Do not provision, restart, scale, or modify Azure Local, AKS Arc, Foundry Local, the model deployment, or portal resources.
  • Do not bypass TLS validation or add --insecure.
  • Ask for explicit confirmation before running an elevated setup, restarting SQL Server, initializing the database, invoking the live model, or generating a live receiving briefing.
  • Never request a secret through chat. If a terminal prompts for one, tell the user to enter it directly in the terminal.

Select a Mode

Infer the mode from the request. If it is unclear, ask the user to choose:

  • prepare: configure a demo VM and database using existing infrastructure.
  • preflight: perform non-destructive readiness checks before presenting.
  • rehearse: guide the locked presentation sequence one beat at a time.
  • troubleshoot: diagnose one failed gate without broad environment changes.

Prepare

  1. Confirm the user is on the intended Windows demo VM and read setup guidance.
  2. Check, without printing sensitive contents:
    • PowerShell is elevated before setup.
    • C:\setup\SJ-SQLAI-admin.kubeconfig exists, or record the alternate path.
    • SQL Server 2025 is running.
    • az, kubectl, kubelogin, curl.exe, sqlcmd, and the .NET 10 SDK are available.
  3. Explain that setup imports a root certificate and restarts SQL Server. After explicit confirmation, run from an elevated PowerShell terminal:

    Set-Location C:\setup
    Set-ExecutionPolicy -Scope Process RemoteSigned
    .\Setup.ps1

    Pass documented overrides only when the user supplied them.

  4. Require exit code 0 and Setup completed successfully.
  5. Ask before invoking the live model. If approved, run C:\setup\test.ps1 and require trusted HTTPS plus HTTP 200 through SQL Server. A successful direct curl call is not sufficient.
  6. From demo1-azure-local, determine whether CaldovaRegionalCare already exists. Use -InitializeDatabase only for an absent database and only after explicit confirmation:

    .\database\Setup-CaldovaDatabase.ps1 -InitializeDatabase

    For an existing database, run the repeat-safe update without that switch.

  7. Run the live AI validation only after separate confirmation:

    .\database\Setup-CaldovaDatabase.ps1 -TestFoundryAi
  8. Build the application. Start app\run.ps1 as a long-running process only when the user asks to launch it.

Preflight

Perform checks without repairing or provisioning:

  1. Confirm the application responds at http://localhost:5099.
  2. Confirm /api/health reports Healthy for CaldovaRegionalCare.
  3. Confirm the MSSQL extension can connect to localhost and database CaldovaRegionalCare.
  4. Confirm the Azure portal is already signed in and the Azure Local VM, AKS cluster, and Azure Virtual Desktop resources are ready to show.
  5. Confirm Phi-3.5 Mini appears in the VS Code model picker. Do not send a clinical prompt.
  6. Report each gate as PASS, FAIL, or NOT CHECKED. Stop on a failed gate and offer the smallest relevant repair from the documented scripts.

Rehearse

Follow the runbook's locked order exactly:

  1. Transfer Center application.
  2. Azure portal: Azure Local VM, AKS, and Azure Virtual Desktop.
  3. VS Code MSSQL connection and ordered database scripts.
  4. Native RegEx extraction.
  5. Approved ai.Skill row.
  6. Transfer-skill procedure, JSON payload, and sys.sp_invoke_external_rest_endpoint.
  7. Phi-3.5 Mini in VS Code Chat.

For each beat, present only its Say, Show, and Land guidance, then wait for the user to continue. Do not execute setup or repair commands during a rehearsal. Do not expose credential rows when showing endpoint metadata.

Troubleshoot

  • Start from the failed gate and its exact error.
  • Read the owning script before proposing a change.
  • Prefer rerunning a documented validation over changing infrastructure.
  • Do not use Connect-AksArc.ps1 unless direct kubeconfig access failed and the user explicitly asks to troubleshoot Arc Cluster Connect.
  • End with the failed command, the evidence found, and the next documented action.

Completion Report

Report the selected mode, checks performed, commands run, gates passed, and any remaining manual step. Never include secret values in the report.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.