Tessl • The management layer for AI agent skills

SkillPublic

airunway-aks-setup

Set up AI Runway on AKS — from bare cluster to running model. Covers cluster verification, controller install, GPU assessment, provider setup, and first deployment. WHEN: "setup AI Runway", "onboard AKS cluster", "install AI Runway", "airunway setup", "deploy model to AKS", "GPU inference on AKS", "KAITO setup on AKS", "run LLM on AKS", "vLLM on AKS", "set up model serving on AKS", "AI Runway controller".

Quality

92%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

88%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-engineered orchestration body: sequenced steps with explicit validation checkpoints, confirmation and cost gates, and concrete error-handling commands. The two minor gaps are redundant MCP/when-to-use sections that could be trimmed, and the two-hop reference chain from step files into shared lookup references.

Suggestions

Delete the "## MCP Tools" section and rely on the Quick Reference table's "MCP tools | None" row — the fact is stated twice.

Drop or compress "## When to Use This Skill" since it restates the frontmatter WHEN triggers nearly verbatim; alternatively differentiate it (e.g., link the aks-gpu-inference boundary decision) instead of repeating.

Surface gpu-profiles.md, model-sizing.md, and powershell-notes.md (or at least the main ones) from the SKILL.md overview so every reference is reachable in one hop instead of only via step files.

Dimension	Reasoning	Score
Conciseness	The body is lean and table-driven with no explanations of concepts Claude already knows, but "MCP tools \| None" in the Quick Reference table is duplicated by a dedicated "## MCP Tools" section, and "When to Use This Skill" restates the frontmatter WHEN triggers nearly verbatim — minor trimmable redundancy matching the 4 anchor rather than the every-token-earns-its-place ideal of 5.	4 / 5
Actionability	Concrete copy-paste commands appear throughout ("kubectl logs -n airunway-system -l control-plane=controller-manager --previous", "az aks get-credentials", "--dtype float16" for bfloat16 errors) alongside exact error→cause→remediation mappings for common cases, fully actionable for an instruction-only skill.	5 / 5
Workflow Clarity	Six steps are clearly sequenced with per-step validation ("Report cluster state at each step: ✓ healthy, ✗ missing/failed"), confirmation gates before installs, an explicit cost gate before GPU provisioning, already-complete skip and skip-to-step resume rules, and an error-recovery table — explicit checkpoints with feedback loops.	5 / 5
Progressive Disclosure	The overview is well-signaled with one row per step linking to real reference files (all verified present), but step files chain a second hop to ../gpu-profiles.md, ../model-sizing.md, and ../powershell-notes.md, slightly exceeding the one-level-deep ideal — good structure with a minor organization gap (4) rather than clear one-level-deep references (5).	4 / 5
	Total	18 / 20 Passed

Description

96%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description: concrete third-person capability list, explicit WHEN clause with abundant natural trigger phrases, and a clearly scoped AKS/AI-Runway niche. The only weakness is that a few inference-oriented triggers could collide with the related Day-2 GPU skill since the description itself does not state that boundary.

Dimension	Reasoning	Score
Specificity	"Set up AI Runway on AKS — from bare cluster to running model. Covers cluster verification, controller install, GPU assessment, provider setup, and first deployment" lists five concrete action phases in third-person voice, matching the comprehensive-coverage anchor; it exceeds the minor-gaps anchor (4) because no setup phase is missing.	5 / 5
Completeness	Both questions are answered explicitly: a clear "what" (the five setup phases) and a dedicated "WHEN:" clause with concrete trigger phrases, exactly matching the 5 anchor; it is above 4 because the "when" is fully explicit, not merely present.	5 / 5
Trigger Term Quality	Eleven explicit triggers including natural user phrasings ("setup AI Runway", "onboard AKS cluster", "run LLM on AKS", "vLLM on AKS", "KAITO setup on AKS") with synonyms covering inference/model serving/LLM; synonym coverage is thorough rather than missing common variations, matching the 5 anchor.	5 / 5
Distinctiveness Conflict Risk	The AI Runway on AKS setup niche is clear, but triggers like "GPU inference on AKS", "run LLM on AKS", and "vLLM on AKS" plausibly overlap with the sibling aks-gpu-inference skill whose boundary is only drawn in the body, matching the minor-overlap-with-closely-related-skills anchor (4) rather than minimal-conflict (5).	4 / 5
	Total	19 / 20 Passed

Validation

81%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 6 deeper-than-1-level	Warning
referenced_paths_exist	Referenced path issues: 6 deeper-than-1-level	Warning

	Total	13 / 16 Passed

Repository: microsoft/GitHub-Copilot-for-Azure
Path: plugins/azure-skills/skills/airunway-aks-setup/SKILL.md
Revision: 156a7dc

Reviewed: 14 days ago

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.

airunway-aks-setup

Table of Contents

Is this your skill?