{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/3090fd76068f42f2bbb697ae669e2cfd\" frameborder=\"0\" width=\"1670\" height=\"1252\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1252,"width":1670,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1252,"thumbnail_width":1670,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/3090fd76068f42f2bbb697ae669e2cfd-d90c4b4f83aa8a53.gif","duration":604.266,"title":"Software Dark Factory PR Review Framework Explained","description":"This Loom explains how Software Dark Factory (SDF) should structure pull requests to make them easier for reviewers and safer for production. It contrasts the unstructured PR review burden with an SDF PR format that includes Motivation, Description, Testing, CI, and a Software Dark Factory specific evidence section covering prompt run context, playbooks applied, and verification details like commands, package installs, and test outcomes. The agent must load SDF governance by referencing a front door framework in the agent file, without mutating the app, to supply review evidence and serve as memory for future agent decisions. The discussion emphasizes maintaining the team quality bar while enabling agentic speed, without forcing changes to engineers tools or IDE."}