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).
Install with Tessl CLI
npx tessl i github:microsoft/github-copilot-for-azure --skill customize88
Quality
85%
Does it follow best practices?
Impact
89%
4.45xAverage score across 3 eval scenarios
Optimize this skill with Tessl
npx tessl skill review --optimize ./plugin/skills/microsoft-foundry/models/deploy-model/customize/SKILL.mdEvaluation results
77%
50%Production Document Processing Pipeline Deployment
PTU deployment sizing and configuration
Live SKU query
20%
100%
SKU quota check
50%
100%
Capacity REST API
30%
100%
PTU capacity min/step
0%
100%
PTU capacity max
0%
100%
PTU default capacity
0%
100%
PTU formula
0%
100%
Priority processing availability
0%
62%
Dynamic quota availability
0%
0%
Version upgrade policy
100%
37%
Deployment name validation
0%
12%
Model format detection
100%
100%
Deployment status polling
50%
100%
Without context: $0.5460 · 3m 10s · 15 turns · 20 in / 11,877 out tokens
With context: $0.7323 · 2m 49s · 23 turns · 1,591 in / 10,517 out tokens
100%
82%Clinical Documentation Assistant — Azure AI Foundry Deployment
Anthropic model deployment via ARM REST API
ARM REST API used
57%
100%
modelProviderData in body
0%
100%
Industry field present
0%
100%
Industry not hardcoded
0%
100%
Industry list presented
0%
100%
Tenant API call
0%
100%
countryCode in payload
0%
100%
organizationName in payload
0%
100%
Capacity set to 1
100%
100%
No RAI policy
0%
100%
No version upgrade policy
0%
100%
Without context: $0.5292 · 2m 59s · 12 turns · 17 in / 12,178 out tokens
With context: $0.5574 · 2m 5s · 22 turns · 1,558 in / 7,802 out tokens
92%
77%Quota-Resilient Deployment Automation for Azure AI Foundry
Cross-region capacity fallback and deployment guards
SKU quota filter
12%
37%
Quota key pattern
0%
100%
Regional capacity REST API
0%
100%
Deployment blocked on failure
66%
100%
Cross-region global endpoint
0%
100%
Cross-region capacity filter
50%
100%
Cross-region sorted descending
0%
100%
AI project listing in fallback region
0%
100%
No-project fallback message
0%
100%
Deployment name validation
20%
100%
Only deployable SKUs shown
0%
50%
Without context: $0.3235 · 1m 35s · 12 turns · 19 in / 6,324 out tokens
With context: $0.6711 · 2m 44s · 21 turns · 1,597 in / 10,393 out tokens
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
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.