CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

cost-optimization

Optimize existing Azure resources and analyze Reservations or Savings Plans. WHEN: "optimize Azure costs", "reduce cloud spending", "rightsize resources", "find idle resources", "orphaned disk", "deleted VM still charged", "public IP still charging", "reservation utilization", "Savings Plan coverage", "commitment recommendation", "why is pay-as-you-go still charged". DO NOT USE FOR: cost spikes, forecasts, pricing estimates, budgets, or governance.

Quality

89%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

The canonical home for this skill is cost-optimization in microsoft/GitHub-Copilot-for-Azure

SKILL.md
Quality
Evals
Security

Quality

Content

82%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-structured, token-efficient instruction-only skill: terse routing overview, a clear recommend-only workflow, and real error-handling guidance. The main gaps are mild — validation checkpoints live outside the numbered sequence, and one bundled reference file is orphaned while service details sit a second reference level deep.

DimensionReasoningScore

Conciseness

At roughly 35 lines the body is lean and assumes competence: no concept explanations, a routing table instead of prose, five terse numbered steps, and a compact error table. The one-line "When to Use This Skill" section mildly duplicates the frontmatter, but every other token earns its place, matching the lean-and-efficient anchor rather than the level-4 'minor instances of over-explanation' anchor.

5 / 5

Actionability

Concrete guidance is present — named tools ("list_benefit_utilization", "list_reservation_transactions", "get_benefit_recommendations"), a specific fallback path ("[API mappings](references/tool-fallback.md)"), and executable error actions ("Retry once; then report the trace ID"). However the optimization row gives only the generic "Cost and Resource Graph tools" without naming them, and no literal commands appear in the body, so it is mostly — not fully — executable guidance with minor gaps.

4 / 5

Workflow Clarity

The five-step sequence is clear and ordered (scope → load workflow → query constraints → separate evidence → recommend only), with error-recovery actions in the error table and validation direction ("Validate queries", "validated server-side projection"). But validation checkpoints are referenced in the MCP Tools section rather than embedded as explicit steps in the sequence (e.g., no 'verify the query validated before executing' step), which matches 'clear sequence with most checkpoints present; minor validation gaps' rather than the fully checkpointed level-5 anchor. The destructive-operation cap does not apply since the skill is recommend-only ("do not delete, resize, purchase, or deploy").

4 / 5

Progressive Disclosure

The body is a genuine overview with a Quick Reference table routing to the two matching workflows plus tool-fallback, and all referenced files exist. Two structural issues keep it below a 5: the chain SKILL.md → optimization.md → services/*.md / resource-graph.md / report-template.md nests content two levels deep, and services/storage-lifecycle.md is not referenced anywhere in the bundle, making it undiscoverable. This matches 'good structure; most content appropriately placed; minor organization gaps' rather than clean one-level-deep referencing.

4 / 5

Total

17

/

20

Passed

Description

90%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 domain scope, an explicit WHEN clause with unusually rich and natural trigger phrases, and a negative boundary that sharply reduces mis-triggering. The only weakness is that the capability statement itself uses two broad verbs, leaving the specificity of actual actions to be inferred from the trigger list.

DimensionReasoningScore

Specificity

The what-clause names the domain and two concrete actions — "Optimize existing Azure resources and analyze Reservations or Savings Plans" — but 'optimize' and 'analyze' are broad verbs with no finer-grained capability breakdown in the statement itself. This matches the anchor for naming a domain with 1-2 concrete actions that are not comprehensive; the finer detail (rightsize, find idle resources) lives only in the trigger list, so it does not reach 'several specific actions' as in the level-4 anchor.

3 / 5

Completeness

Explicitly answers what ("Optimize existing Azure resources and analyze Reservations or Savings Plans") and when ("WHEN:" followed by concrete trigger phrases), plus an explicit negative boundary ("DO NOT USE FOR: cost spikes, forecasts, pricing estimates, budgets, or governance"). This clearly matches the anchor requiring both what AND when with concrete trigger phrases; the level-4 anchor ('when could be more explicit') is a weaker fit given the dedicated WHEN clause.

5 / 5

Trigger Term Quality

Eleven natural trigger phrases including colloquial and question forms users would actually say: "optimize Azure costs", "rightsize resources", "orphaned disk", "deleted VM still charged", "why is pay-as-you-go still charged", "Savings Plan coverage". Coverage spans synonyms, symptoms, and question phrasing with no obvious common variant missing, matching the comprehensive-coverage anchor; the level-4 anchor ('a few natural terms missing') fits worse.

5 / 5

Distinctiveness Conflict Risk

A clear niche (Azure waste/rightsizing and commitment analysis) with distinct triggers and an explicit DO NOT USE FOR exclusion list that prevents firing on adjacent cost-management intents (spikes, forecasts, budgets, governance). Minimal conflict risk, matching the clear-niche anchor rather than the level-4 'minor overlap risk' anchor.

5 / 5

Total

18

/

20

Passed

Validation

100%

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

microsoft/azure-skills

Path

.github/plugins/azure-cost/skills/cost-optimization/SKILL.md

Reviewed

Table of Contents

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.