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.
In this repository18 other files it gives its agents.
- biomedical-evidence-review-v2Skill
- analyze-readiness-at-scaleSkill
- evaluate-azure-migration-assessmentSkill
- evaluate-offline-migration-readinessSkill
- generate-migration-prerequisite-planSkill
- get-migration-assessmentSkill
- recommend-migration-pathSkill
- recommend-sku-sizingSkill
- run-migration-assessmentSkill
- sql-backup-restore-to-azure-sql-vm-migrationSkill
- sql-bacpac-to-azure-sql-db-migrationSkill
- sql-server-to-sql-mi-lrs-migrationSkill
- validate-post-migration-dataSkill
- build-caldova-evidence-agentSkill
- demo-caldova-evidence-agentSkill
- run-caldova-evidence-agentSkill
- run-handsfree-indexingSkill
- keynote-demo-appSkill
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
- Very specific: line 17
Never display or read into chat the kubeconfig, Kubernetes API key, database scoped credential secret, database master-key password, or a …
- Distinctive: line 20
Never retrieve a Kubernetes Secret directly for inspection.
- Distinctive: line 22
Do not provision, restart, scale, or modify Azure Local, AKS Arc, Foundry Local, the model deployment, or portal resources.
- Very specific: line 24
Do not bypass TLS validation or add --insecure.
- Distinctive: line 25
Ask for explicit confirmation before running an elevated setup, restarting SQL Server, initializing the database, invoking the live model, …
- Distinctive: line 28
Never request a secret through chat.
- Distinctive: line 44
Check, without printing sensitive contents:
- Very specific: line 61
Ask before invoking the live model.
- Distinctive: line 73
Run the live AI validation only after separate confirmation:
- Distinctive: line 99
Follow the runbook's locked order exactly:
- Distinctive: line 116
Start from the failed gate and its exact error.
- Distinctive: line 117
Read the owning script before proposing a change.
- Distinctive: line 118
Prefer rerunning a documented validation over changing infrastructure.
- Very specific: line 119
Do not use Connect-AksArc.ps1 unless direct kubeconfig access failed and the user explicitly asks to troubleshoot Arc Cluster Connect.
≈ “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.
Files
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
- Confirm the user is on the intended Windows demo VM and read setup guidance.
- Check, without printing sensitive contents:
- PowerShell is elevated before setup.
C:\setup\SJ-SQLAI-admin.kubeconfigexists, or record the alternate path.- SQL Server 2025 is running.
az,kubectl,kubelogin,curl.exe,sqlcmd, and the .NET 10 SDK are available.
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.ps1Pass documented overrides only when the user supplied them.
- Require exit code 0 and
Setup completed successfully. - Ask before invoking the live model. If approved, run
C:\setup\test.ps1and require trusted HTTPS plus HTTP 200 through SQL Server. A successful directcurlcall is not sufficient. From
demo1-azure-local, determine whetherCaldovaRegionalCarealready exists. Use-InitializeDatabaseonly for an absent database and only after explicit confirmation:.\database\Setup-CaldovaDatabase.ps1 -InitializeDatabaseFor an existing database, run the repeat-safe update without that switch.
Run the live AI validation only after separate confirmation:
.\database\Setup-CaldovaDatabase.ps1 -TestFoundryAi- Build the application. Start
app\run.ps1as a long-running process only when the user asks to launch it.
Preflight
Perform checks without repairing or provisioning:
- Confirm the application responds at
http://localhost:5099. - Confirm
/api/healthreportsHealthyforCaldovaRegionalCare. - Confirm the MSSQL extension can connect to
localhostand databaseCaldovaRegionalCare. - Confirm the Azure portal is already signed in and the Azure Local VM, AKS cluster, and Azure Virtual Desktop resources are ready to show.
- Confirm Phi-3.5 Mini appears in the VS Code model picker. Do not send a clinical prompt.
- Report each gate as
PASS,FAIL, orNOT CHECKED. Stop on a failed gate and offer the smallest relevant repair from the documented scripts.
Rehearse
Follow the runbook's locked order exactly:
- Transfer Center application.
- Azure portal: Azure Local VM, AKS, and Azure Virtual Desktop.
- VS Code MSSQL connection and ordered database scripts.
- Native RegEx extraction.
- Approved
ai.Skillrow. - Transfer-skill procedure, JSON payload, and
sys.sp_invoke_external_rest_endpoint. - 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.ps1unless 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.
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
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.

