customize
Interactive guided deployment flow for Azure OpenAI models with full customization control. Step-by-step selection of model version, SKU (GlobalStandard/Standard/ProvisionedManaged), capacity, RAI policy (content filter), and advanced options (dynamic quota, priority processing, spillover). USE FOR: custom deployment, customize model deployment, choose version, select SKU, set capacity, configure content filter, RAI policy, deployment options, detailed deployment, advanced deployment, PTU deployment, provisioned throughput. DO NOT USE FOR: quick deployment to optimal region (use preset).
84
78%
Does it follow best practices?
Impact
89%
4.45xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./plugin/skills/microsoft-foundry/models/deploy-model/customize/SKILL.mdQuality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is an excellent skill description that covers all key dimensions thoroughly. It provides specific concrete actions, comprehensive trigger terms users would naturally use, explicit 'USE FOR' and 'DO NOT USE FOR' clauses for clear selection guidance, and strong distinctiveness through negative trigger guidance that disambiguates it from a related quick-deployment skill.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: step-by-step selection of model version, SKU types (GlobalStandard/Standard/ProvisionedManaged), capacity, RAI policy (content filter), and advanced options (dynamic quota, priority processing, spillover). Very detailed and concrete. | 3 / 3 |
Completeness | Clearly answers both 'what' (interactive guided deployment flow for Azure OpenAI models with full customization control) and 'when' (explicit USE FOR and DO NOT USE FOR clauses with detailed trigger scenarios). The negative trigger guidance is a bonus for disambiguation. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms including 'custom deployment', 'choose version', 'select SKU', 'set capacity', 'configure content filter', 'RAI policy', 'advanced deployment', 'PTU deployment', 'provisioned throughput'. These are terms users would naturally use when requesting these capabilities. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with a clear niche (custom/advanced Azure OpenAI deployment) and explicitly distinguishes itself from a 'quick deployment to optimal region' skill with a DO NOT USE FOR clause, minimizing conflict risk. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
57%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is well-structured as an overview document with clear progressive disclosure to reference files, but it sacrifices actionability by deferring all executable content to external files. The workflow sequence is clearly laid out but lacks inline validation commands. The document could be more concise by trimming the comparison table and 'When to Use' checklist, which largely duplicate the description metadata.
Suggestions
Include at least one complete executable example inline (e.g., the core deployment CLI command with all parameters filled in) so the SKILL.md is actionable on its own without loading references.
Trim the 'When to Use This Skill' checklist and the customize-vs-preset comparison table — this information is better suited for the skill description/metadata rather than consuming tokens in the body.
Add an explicit validation checkpoint inline for the deployment phase (e.g., 'After deploy, verify with: `az cognitiveservices account deployment show ...` — confirm status is Succeeded before reporting completion').
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content includes useful tables and structured information but is verbose in places — the 'When to Use This Skill' section with checkmarks, the comparison table between customize vs preset, and the Quick Reference table all add bulk that Claude doesn't need. The phase summaries table is reasonably efficient but the overall document is longer than necessary. | 2 / 3 |
Actionability | The skill provides some concrete CLI commands in the troubleshooting section and mentions specific commands like `az cognitiveservices account deployment create`, but the actual workflow implementation is deferred entirely to `references/customize-workflow.md`. The SKILL.md itself contains no executable deployment code — it describes what to do rather than showing how. | 2 / 3 |
Workflow Clarity | The 14-phase workflow is clearly sequenced with a good overview and the phase summaries table provides useful structure. However, the actual validation checkpoints and feedback loops are only described at a summary level (e.g., 'cross-region fallback if no capacity') with implementation deferred to the reference file. The capacity validation blocking deployment is mentioned but the actual validation commands are not shown inline. | 2 / 3 |
Progressive Disclosure | The skill effectively uses a summary-level SKILL.md that clearly signals one-level-deep references: `references/customize-workflow.md` for full scripts and `references/customize-guides.md` for SKU guides and advanced topics. The ⚠️ MUST READ callout is well-placed, and the content split between overview and reference files is appropriate. | 3 / 3 |
Total | 9 / 12 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 — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- microsoft/github-copilot-for-azure
- Commit
915f809
- 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.