skypilot-multi-cloud-orchestration

Multi-cloud orchestration for ML workloads with automatic cost optimization. Use when you need to run training or batch jobs across multiple clouds, leverage spot instances with auto-recovery, or optimize GPU costs across providers.

Quality

80%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./bundled/skills/skypilot-multi-cloud-orchestration/SKILL.md

Quality

Content

76%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 highly actionable, well-organized reference rich in executable examples and real bundle references. The main gap is workflow safety: destructive and batch operations lack the explicit validation checkpoints the rubric requires for a higher workflow-clarity score.

Suggestions

Add explicit validation checkpoints before destructive actions (e.g., 'Run sky status first; confirm the cluster name before sky down') and a validate-fix-retry loop around batch sweeps.

Move the long accelerator lists and duplicated envs/secrets blocks into references/advanced-usage.md to tighten the overview and lift progressive disclosure.

Cut the redundant environment-variable example (secrets vs. envs) to one canonical pattern to improve conciseness.

Dimension	Reasoning	Score
Conciseness	Largely lean — code blocks, tables, and brief bullets dominate with little concept-explanation padding — but a few sections (repeated envs/secrets, expansive accelerator lists) could be trimmed, so it sits just below fully lean.	4 / 5
Actionability	Provides copy-paste-ready, executable YAML and bash covering the common cases (launch, distributed training, spot recovery, serving, sweeps), with specific commands and complete configuration rather than pseudocode.	5 / 5
Workflow Clarity	Workflows are sequenced into clear sections, but the skill drives destructive/batch operations (sky down, multi-job sweeps) and lacks explicit validation checkpoints or validate-fix-retry feedback loops beyond a single --dryrun mention, so the destructive-operations cap applies.	3 / 5
Progressive Disclosure	Good organization with clearly signaled, verified one-level-deep references to references/advanced-usage.md and references/troubleshooting.md, but a large volume of inlined YAML examples keeps it just short of an ideal overview-plus-references split.	4 / 5
	Total	16 / 20 Passed

Description

83%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, well-structured description that clearly states capabilities and explicit use-when triggers with natural keywords. Its only real weakness is the second-person voice, which is explicitly penalized and prevents a top specificity score.

Suggestions

Switch to third-person voice (e.g., 'Use when running training or batch jobs...') to recover the specificity point lost to the second-person penalty.

Add a few synonymous trigger terms (e.g., 'cheapest GPU', 'preemptible instances') to push trigger-term coverage toward comprehensive.

Tighten distinctiveness by naming the specific differentiator (SkyPilot) so it cannot collide with generic multi-cloud skills.

Dimension	Reasoning	Score
Specificity	Names multiple concrete actions ('run training or batch jobs across multiple clouds, leverage spot instances with auto-recovery, or optimize GPU costs across providers'), which would warrant a 5, but the second-person 'Use when you need to run' triggers the -1 voice penalty.	4 / 5
Completeness	Explicitly answers both what ('Multi-cloud orchestration for ML workloads with automatic cost optimization') and when ('Use when you need to run training or batch jobs across multiple clouds, leverage spot instances with auto-recovery, or optimize GPU costs across providers') with concrete trigger phrases.	5 / 5
Trigger Term Quality	Good natural keyword coverage ('training', 'batch jobs', 'multiple clouds', 'spot instances', 'auto-recovery', 'GPU costs'), but a few common variations or synonyms are missing, keeping it just below comprehensive.	4 / 5
Distinctiveness Conflict Risk	The multi-cloud ML/GPU/spot niche is mostly distinct with minimal conflict risk, though 'multi-cloud orchestration' is broad enough to overlap with adjacent infra skills.	4 / 5
	Total	17 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (510 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: foryourhealth111-pixel/Vibe-Skills
Path: bundled/skills/skypilot-multi-cloud-orchestration/SKILL.md
Commit: ddcaa2a

Reviewed: 8 days ago

Table of Contents

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.