<?xml version="1.0" encoding="UTF-8"?><oembed><type>video</type><version>1.0</version><html>&lt;iframe src=&quot;https://www.loom.com/embed/fa6ad4aef19f46cca9315503f34cc033&quot; frameborder=&quot;0&quot; width=&quot;1920&quot; height=&quot;1440&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>1440</height><width>1920</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>1440</thumbnail_height><thumbnail_width>1920</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/fa6ad4aef19f46cca9315503f34cc033-22cfa3d435c1773a.gif</thumbnail_url><duration>64.833</duration><title>Semantic Gateway Stops Destructive SQL Agents 🛡️</title><description>Hi, I’m building Semantic Gateway, a firewall to stop autonomous AI agents from dropping your database tables. In this Loom, I show a destructive SQL action being captured at the edge, with the JSON intents shared to the dashboard. The agent is then paused and waiting for the user to take action. On the right side, I demonstrate what happens when the user responds from the security dashboard.</description></oembed>