<?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/eb484ed3765443fb963f8f5634c4f7a2&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/eb484ed3765443fb963f8f5634c4f7a2-4f6ee8a717621079.gif</thumbnail_url><duration>741.967203</duration><title>Secureless AI Security Agent for GitHub</title><description>Hi, I am Krishna Singh, and I walked you through Secureless AI, my security agent that scans GateHub repositories for vulnerabilities and holds a live conversation with you about what it finds and how to fix it. It first decides which files are most likely to contain issues, then fetches up to 5 files in parallel and checks the top 10 vulnerabilities per file, using Python asyncio to keep it fast. Results are stored in memory, then you can chat with the agent for patches and explanations. It is built with FastAPI and runs in Docker, using Gemini 2.0 Flash and the Gemini SDK, plus an async GitHub client. I did not ask you to take any action, but I demonstrated how scanning works via wrapper URL, GitHub token, branch, and scan ID.</description></oembed>