azure-aigateway
Configure Azure API Management as an AI Gateway for AI models, MCP tools, and agents. WHEN: semantic caching, token limit, content safety, load balancing, AI model governance, MCP rate limiting, jailbreak detection, add Azure OpenAI backend, add AI Foundry model, test AI gateway, LLM policies, configure AI backend, token metrics, AI cost control, convert API to MCP, import OpenAPI to gateway.
95
93%
Does it follow best practices?
Impact
95%
1.72xAverage score across 3 eval scenarios
Risky
Do not use without reviewing
Quality
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 a strong skill description that clearly defines its scope (Azure API Management as AI Gateway), provides an explicit WHEN clause with extensive trigger terms, and occupies a distinct niche. The trigger terms are well-chosen and cover both high-level concepts (AI cost control, AI model governance) and specific tasks (add Azure OpenAI backend, import OpenAPI to gateway). The description is comprehensive without being verbose.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description lists multiple specific concrete actions: configuring Azure API Management as an AI Gateway, semantic caching, token limiting, content safety, load balancing, jailbreak detection, adding Azure OpenAI backends, converting APIs to MCP, importing OpenAPI specs, and more. | 3 / 3 |
Completeness | Clearly answers both 'what' (Configure Azure API Management as an AI Gateway for AI models, MCP tools, and agents) and 'when' with an explicit 'WHEN:' clause listing numerous specific trigger scenarios. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms users would say: 'semantic caching', 'token limit', 'content safety', 'load balancing', 'jailbreak detection', 'Azure OpenAI backend', 'AI Foundry model', 'MCP rate limiting', 'convert API to MCP', 'import OpenAPI to gateway'. These are terms practitioners would naturally use. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with a clear niche: Azure API Management as an AI Gateway. The combination of Azure APIM, AI gateway policies, MCP tools, and specific features like jailbreak detection and AI Foundry models makes it very unlikely to conflict with other skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
87%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured skill that efficiently covers Azure AI Gateway configuration with concrete CLI commands, clear policy references, and excellent progressive disclosure. The main weakness is the lack of explicit validation steps in multi-step workflows like adding backends and applying policies, which could lead to silent failures. The troubleshooting table is a nice touch but doesn't substitute for inline verification steps.
Suggestions
Add validation commands after key steps (e.g., verify backend creation with `az apim backend show`, verify role assignment with `az role assignment list`) to create feedback loops in the 'Add AI Backend' workflow.
Add a brief verification step after 'Apply AI Governance Policy' to confirm the policy is active (e.g., a test request showing the policy in effect).
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is lean and efficient. It avoids explaining what APIM is or how AI gateways work conceptually, instead jumping straight to actionable commands, policy references, and task-oriented sections. Every section earns its place. | 3 / 3 |
Actionability | Provides fully executable Azure CLI commands for getting gateway details, testing endpoints, and adding backends. The curl example is copy-paste ready with clear placeholders, and policy ordering gives specific, concrete guidance. | 3 / 3 |
Workflow Clarity | The 'Add AI Backend' section has clear sequential steps (discover, create, grant access), and the policy ordering is well-defined. However, there are no explicit validation checkpoints or feedback loops — e.g., no step to verify the backend was created successfully or that the role assignment took effect before proceeding. | 2 / 3 |
Progressive Disclosure | Excellent progressive disclosure with a concise overview in the main file and well-signaled one-level-deep references to policies.md, patterns.md, troubleshooting.md, and SDK references. The quick reference table links directly to specific sections within reference files. | 3 / 3 |
Total | 11 / 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
a46a937
- 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.