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

—

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Advisory

Suggest reviewing before use

Quality

Content

77%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-organized routing entry point with concrete validation commands, a clear sequenced workflow, and explicit confirmation checkpoints. Its main weaknesses are redundant restatements of the routing logic and progressive-disclosure gaps where referenced sub-skills do not exist in the bundle while existing scripts go unreferenced.

Suggestions

Collapse the overlapping routing logic — keep either the 'Intent Detection' tree or the 'Routing Rules' table, and fold 'Multi-Mode Chaining' into the survivor rather than restating the same decisions three ways.

Ship or stub the referenced sub-skills (preset/SKILL.md, customize/SKILL.md, capacity/SKILL.md) so the signaled one-level-deep navigation resolves, or note their absence explicitly.

Reference the existing scripts/generate_deployment_url.sh and .ps1 from the body (e.g., when surfacing a post-deployment portal link) so the bundle files are not orphaned.

Dimension	Reasoning	Score
Conciseness	The body is mostly lean and assumes Claude's competence (no basic-concept explanation), but the routing logic is restated three times — the ASCII 'Intent Detection' tree, the 'Routing Rules' table, and the 'Multi-Mode Chaining' section — so it could be tightened; not a 1 because there is no padded concept explanation, and not a 3 because of the redundant decision-logic representations.	2 / 3
Actionability	Provides copy-paste-ready executable commands ('az cognitiveservices model list ...', 'az cognitiveservices usage list ...') plus concrete keyword-based routing rules and a literal confirmation dialog template; not a 2 because the guidance is executable and specific rather than pseudocode or abstract.	3 / 3
Workflow Clarity	Sequences project resolution (env var → prompt → query), a mandatory confirmation step with user feedback, and pre-deployment validation (SKU support + quota) with explicit 'Only present options that pass both checks' checkpoints; not a 2 because validation and the feedback loop are explicit rather than implied.	3 / 3
Progressive Disclosure	The overview is well structured with one-level-deep, clearly signaled sub-skill references (preset/SKILL.md, customize/SKILL.md, capacity/SKILL.md) in a Quick Reference table, but those referenced directories are absent from the bundle and the existing scripts/ files (generate_deployment_url.sh/.ps1) are not referenced anywhere; not a 3 because the signaled navigation targets are missing and present bundle files are orphaned.	2 / 3
	Total	10 / 12 Passed

Description

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.

A high-quality, third-person description that concretely enumerates capabilities, supplies natural trigger terms, explicitly states when to use it, and proactively distinguishes itself from adjacent skills via a DO NOT USE FOR block. It is concise and free of vague fluff.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'Handles quick preset deployments, fully customized deployments (version/SKU/capacity/RAI policy), and capacity discovery across regions and projects' — matching the anchor for several specific concrete actions; not a 2 because it goes beyond naming the domain into enumerated, parameterized actions.	3 / 3
Completeness	Explicitly answers what ('Handles quick preset deployments...capacity discovery') and when via the 'USE FOR:' trigger clause plus a 'DO NOT USE FOR:' negative-trigger block, satisfying the both-what-and-when anchor; not a 2 because the when is explicit, not merely implied.	3 / 3
Trigger Term Quality	The 'USE FOR:' list (deploy model, deploy gpt, create deployment, set up model, find capacity, where can I deploy, best region for model, capacity analysis) gives broad coverage of phrasings a user would naturally say; not a 2 because common variations are well represented rather than partial.	3 / 3
Distinctiveness Conflict Risk	A clear Azure OpenAI model-deployment niche with distinct triggers and a 'DO NOT USE FOR:' block redirecting overlapping tasks (listing deployments to the MCP tool, agent/project creation to their own skills), making wrong-skill conflicts unlikely; not a 2 because the negative triggers explicitly carve out adjacent skills.	3 / 3
	Total	12 / 12 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
Commit: 2cd48ca

Reviewed: about 8 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.