Skills airunway-aks-setup
๐Ÿ“ฆ

airunway-aks-setup

v1.0.1 Content revision r2 Medium Risk โš™๏ธ External commands๐Ÿ“ Filesystem access

Set Up AI Runway on AKS

AI Runway setup requires coordinated cluster, GPU, provider, and model checks. This skill guides each AKS stage with confirmation before installation or deployment.

Supports: Claude Codex Code(CC)
๐Ÿ“Š 71 Adequate

Install with my Agent

Copy this request to your Agent. It includes the canonical Skill page and manifest.

Agent request
Review the Skillstore skill "airunway-aks-setup" from https://skillstore.io/skills/microsoft-airunway-aks-setup.md and its manifest at https://skillstore.io/api/skills/microsoft-airunway-aks-setup/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.

Your Agent should still show its plan and request any confirmation required by the security policy.

Agent-readable resources

Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.

Test it

Using "airunway-aks-setup". Verify my AKS cluster and recommend a starter model.

Expected outcome:

  • Cluster context is reachable with three nodes and one 24 GB L4 GPU.
  • Float16 and bfloat16 are supported. KAITO with an 8B starter model fits the available GPU memory.
  • No changes were made. Controller installation requires confirmation.

Using "airunway-aks-setup". Resume at provider setup and deploy a CPU-only model.

Expected outcome:

  • KAITO with llama.cpp is recommended for the CPU-only cluster.
  • The provider is registered and ready. The selected model deployment is still downloading weights.
  • Next check: wait for Ready status, then retrieve the endpoint and run the smoke test.

Security Audit

Medium Risk
v5 โ€ข 7/23/2026 Open versioned report

All 46 static findings are false positives from Markdown syntax, bundled relative links, or expected administrative command examples. Mutating operations require user confirmation, and no prompt injection or malicious intent was found. A separate medium risk remains because the PowerShell flow stores a token at a predictable plaintext temporary path.

11
Files scanned
619
Lines analyzed
0
Review items
0
False positives ignored

Confirmed security concerns (1)

Medium
Predictable Plaintext Token File
The PowerShell flow writes the Hugging Face access token to a predictable plaintext path before creating the Kubernetes secret. Another local process under the same account could read or replace that file before cleanup.
The fixed path and plaintext write are explicit, although the file is removed after kubectl creates the secret.
Audited by: codex View Audit History โ†’
Share & cite this report

Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.

Open versioned report
Security Assessment

Copy report link

https://skillstore.io/skills/microsoft-airunway-aks-setup/audits/5?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown badge

[![Skillstore security assessment](https://skillstore.io/badges/skills/microsoft-airunway-aks-setup/security.svg)](https://skillstore.io/skills/microsoft-airunway-aks-setup?utm_source=security_passport_badge)

HTML badge

<a href="https://skillstore.io/skills/microsoft-airunway-aks-setup?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/microsoft-airunway-aks-setup/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Embed card

<iframe src="https://skillstore.io/embed/skills/microsoft-airunway-aks-setup.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>
Academic citations (APA ยท BibTeX ยท CFF)

APA citation

microsoft. (2026). airunway-aks-setup security audit report (audit version 5) [Author version 1.0.1]. Skillstore. https://skillstore.io/skills/microsoft-airunway-aks-setup/audits/5

BibTeX citation

@techreport{microsoft-microsoft-airunway-aks-setup-2026, author = {microsoft}, title = {airunway-aks-setup security audit report (audit version 5)}, institution = {Skillstore}, year = {2026}, number = {5}, url = {https://skillstore.io/skills/microsoft-airunway-aks-setup/audits/5}, note = {Author version 1.0.1} }

CITATION.cff

cff-version: 1.2.0 message: "If you use this Skill, cite its author and this versioned security audit report." title: "airunway-aks-setup security audit report (audit version 5)" version: "1.0.1" type: report authors: - name: "microsoft" date-released: "2026-07-23" url: "https://skillstore.io/skills/microsoft-airunway-aks-setup/audits/5" identifiers: - type: other value: "skillstore:microsoft-airunway-aks-setup:audit:5" description: "Skillstore immutable audit report identifier"

Skillstore Score

Why this score Evidence Confidence: High
50
Architecture
100
Maintainability
87
Content
68
Community
91
Spec Compliance

What You Can Build

Bootstrap an existing AKS cluster

Verify the cluster, install AI Runway components with confirmation, and deploy a compatible starter model.

Match models to GPU capacity

Assess detected accelerators, identify compatibility limits, and select a provider and model that fit available VRAM.

Resume or recover setup

Start from a named setup step, inspect component status, and follow targeted readiness or rollback guidance.

Try These Prompts

Verify an AKS cluster
Verify my current AKS cluster for AI Runway. Report tools, context, nodes, GPUs, and blockers. Do not install anything.
Complete guided setup
Set up AI Runway on my existing AKS cluster. Stop before each installation or deployment and explain the expected cluster changes.
Choose and deploy a model
Assess my GPU capacity, recommend a compatible provider and starter model, then deploy after confirmation. Monitor readiness and explain any failure.
Resume with an advanced provider
Resume at step 4 on my AKS cluster. Compare Dynamo and KubeRay for high-throughput serving, verify the selected image, and request confirmation before deployment.

Best Practices

  • Verify the active kubeconfig context and namespace before every mutating command.
  • Review GPU costs, model licenses, image sources, and expected cluster changes before confirmation.
  • Use interactive secret entry, least-privilege access, and immediate temporary file cleanup for gated model tokens.

Avoid

  • Do not skip prerequisite or readiness checks when resuming from a later step.
  • Do not place Hugging Face tokens in prompts, shell history, manifests, or shared logs.
  • Do not select models from weight size alone; reserve memory for cache and runtime overhead.

Frequently Asked Questions

Does this skill create an AKS cluster?
No. It requires an existing cluster and working kubeconfig. Cluster provisioning belongs in a separate Azure Kubernetes workflow.
Can it work without GPUs?
Yes. It recommends a small quantized model with KAITO and llama.cpp for CPU-only inference.
Which inference providers are supported?
The guidance covers KAITO, Dynamo, and KubeRay. KAITO is the default for standard inference and CPU-only clusters.
Will it change my cluster automatically?
It performs status checks first and requires confirmation before controller, provider, secret, or model deployment actions.
How are gated model tokens handled?
The workflow reads tokens interactively and creates a Kubernetes secret. Review the PowerShell temporary file warning before use.
Can setup resume after a failed step?
Yes. Provide a step number to resume, then verify assumed prior state and use the rollback guidance when needed.

Developer Details

Author

microsoft

License

MIT

Author version

v1.0.1

Skillstore revision

r2

Ref

ebdfe608f5de2b66ff37ab4af12af8ac4f5e8006

Maintenance freshness

7/25/2026

Usage

6 downloads ยท 54 views

File structure

๐Ÿ“ references/

๐Ÿ“„ gpu-profiles.md

๐Ÿ“„ model-sizing.md

๐Ÿ“„ powershell-notes.md

๐Ÿ“ steps/

๐Ÿ“„ step-1-verify.md

๐Ÿ“„ step-2-controller.md

๐Ÿ“„ step-3-gpu.md

๐Ÿ“„ step-4-provider.md

๐Ÿ“„ step-5-deploy.md

๐Ÿ“„ step-6-summary.md

๐Ÿ“„ troubleshooting.md

๐Ÿ“„ SKILL.md

More from microsoft

View all
View all