<?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/96e206ad8dea4aee9d2c5ea69d9fe5c3&quot; frameborder=&quot;0&quot; width=&quot;1280&quot; height=&quot;960&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>960</height><width>1280</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>960</thumbnail_height><thumbnail_width>1280</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/96e206ad8dea4aee9d2c5ea69d9fe5c3-7673f7bf89ed2fb3.gif</thumbnail_url><duration>300.168</duration><title>Jira Coding Agent Workflow Demo, Simple vs High Risk</title><description>I built a Jira coding agent demo that watches for new Jira ticket creation and autonomously makes code changes in a React repo. For a simple ticket, I cloned the repo, did RAG indexing, captured before and after screenshots, ran tests, and successfully created a PR showing updates in app.js and app.test.js. Then I created a high risk ticket, and the agent classified it as higher risk, presented a plan for review, and only after approval it implemented stateful click counter logic and created the PR. No action was requested from viewers in this Loom.</description></oembed>