deploy-model

Unified Azure OpenAI model deployment skill with intelligent intent-based routing. Handles quick preset deployments, fully customized deployments (version/SKU/capacity/RAI policy), and capacity discovery across regions and projects. USE FOR: deploy model, deploy gpt, create deployment, model deployment, deploy openai model, set up model, provision model, find capacity, check model availability, where can I deploy, best region for model, capacity analysis. DO NOT USE FOR: listing existing deployments (use foundry_models_deployments_list MCP tool), deleting deployments, agent creation (use agent/create), project creation (use project/create).

Quality

85%

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

Quality

Content

71%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 routing skill with concrete validation commands and clear sequencing, weakened by a progressive-disclosure mismatch: the body points to sub-skills and external docs that are absent from the bundle while leaving the bundled scripts unreferenced, so the central deploy action is not self-contained. Conciseness and workflow clarity are strong but not maximal due to minor padding and an implicit error-feedback loop.

Suggestions

Create the referenced sub-skills (preset/SKILL.md, customize/SKILL.md, capacity/SKILL.md) or inline their core deploy steps so the terminal action is executable from SKILL.md alone.

Reference the existing generate_deployment_url.sh / .ps1 scripts from the body (e.g., in a post-deployment 'view in portal' step) so bundled files are not orphaned.

Trim mock data (sample project lists, the 'Reason' routing column) and make the validation-failure path explicit ('if either check fails, do not present the option; inform the user why') to lift conciseness and workflow clarity toward 5.

Dimension	Reasoning	Score
Conciseness	Largely lean — routing tables, an ASCII intent tree, and short validation blocks — but it carries some trimmable padding such as mock project lists ('project-alpha (rg-alpha)...') and a redundant 'Reason' column; it stops short of the every-token-earns-its-place bar of a 5.	4 / 5
Actionability	Provides concrete, executable validation commands with exact JSON paths ('az cognitiveservices model list ...', '.model.skus[].name', 'OpenAI.<SKU>.<model-name>'), but the actual deployment action is delegated to sub-skills that are not present in the bundle, and the bundled scripts are never referenced, leaving the core execute step not copy-paste ready.	4 / 5
Workflow Clarity	Sequences project resolution → confirmation → SKU/quota validation → mode routing with an explicit 'Only present options that pass both checks' checkpoint, satisfying the validation requirement for a deployment operation; the terminal deploy step dangles into missing sub-skills and the validate-fail feedback loop is only implicit, keeping it below a 5.	4 / 5
Progressive Disclosure	The top-level structure and deferral pattern (sub-skills, quota/foundry links) are sensible, but judged against the actual bundle: the referenced sub-skills (preset/, customize/, capacity/) do not exist, while the scripts that do exist (generate_deployment_url.sh/.ps1) are orphaned and never linked — broken references plus an unreferenced real file signal a structure mismatch rather than the clean one-level split of a 4.	3 / 5
	Total	15 / 20 Passed

Description

100%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.

The description is exemplary: it states concrete capabilities, supplies a rich set of natural trigger terms, explicitly marks when to use it and when not to, and sharply distinguishes itself from neighboring skills. It hits the top anchor on every dimension.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions across three deployment modes — 'quick preset deployments', 'fully customized deployments (version/SKU/capacity/RAI policy)', and 'capacity discovery across regions and projects' — giving comprehensive coverage rather than the minor gaps of a 4.	5 / 5
Completeness	Explicitly answers both 'what' (the three deployment modes) and 'when' (concrete 'USE FOR:' triggers plus a 'DO NOT USE FOR:' boundary), matching the anchor that requires concrete trigger phrases.	5 / 5
Trigger Term Quality	The 'USE FOR:' clause packs natural phrases users would actually say — 'deploy model', 'deploy gpt', 'find capacity', 'where can I deploy', 'best region for model' — covering synonyms and variations comprehensively, beyond the 'few missing' threshold of a 4.	5 / 5
Distinctiveness Conflict Risk	A clear Azure OpenAI deployment niche with distinct triggers, and the 'DO NOT USE FOR:' clause actively disambiguates from related skills (list/delete deployments, agent creation, project creation), minimizing conflict risk.	5 / 5
	Total	20 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 6 missing, 2 suspicious	Warning

	Total	15 / 16 Passed

Repository: microsoft/azure-skills
Path: .github/plugins/azure-skills/skills/microsoft-foundry/models/deploy-model/SKILL.md
Commit: 16dc8c5

Reviewed: about 12 hours 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.