<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>AI Gateway on Top AI Skills</title><link>https://topaiskills.com/tags/ai-gateway/</link><description>Recent content in AI Gateway on Top AI Skills</description><generator>Hugo</generator><language>en</language><lastBuildDate>Thu, 17 Sep 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://topaiskills.com/tags/ai-gateway/index.xml" rel="self" type="application/rss+xml"/><item><title>Azure AI Gateway: 6 Steps to Govern OpenAI Calls</title><link>https://topaiskills.com/tutorials/guides/azure-aigateway-procedural/</link><pubDate>Thu, 17 Sep 2026 00:00:00 +0000</pubDate><guid>https://topaiskills.com/tutorials/guides/azure-aigateway-procedural/</guid><description>&lt;p&gt;The moment your app starts calling Azure OpenAI directly, three things escape you: cost, rate limits, and what content is actually flowing through. Microsoft&amp;rsquo;s azure-aigateway skill (v3.2.1) fixes that by configuring Azure API Management as an AI gateway — a control plane sitting between your callers and the model. I read the SKILL.md plus all four references (patterns, policies, troubleshooting, auth best practices) in &lt;a href="https://github.com/microsoft/azure-skills" rel="nofollow" target="_blank"&gt;microsoft/azure-skills&lt;/a&gt;, then scraped skills.sh on 2026-09-17: 585,092 installs, rank 57, lockstepping with azure-kusto two spots below. The procedure below is the skill&amp;rsquo;s own Pattern 1, sequenced.&lt;/p&gt;</description></item></channel></rss>