{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/330d2fa4eb4f4cc6801c880073edf178\" frameborder=\"0\" width=\"1280\" height=\"960\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":960,"width":1280,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":960,"thumbnail_width":1280,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/330d2fa4eb4f4cc6801c880073edf178-5f14f7edebba2d15.gif","duration":872.555,"title":"PR Review Agent Demo and Workflow Overview","description":"Hi, welcome to the demo of our PR Review Agent. I built this app to review the latest pull request for one or more repos, calculate confidence and metascore inputs, run deterministic checks and tests, then use agents to extract issues, generate fixes, and summarize everything for the project manager. The app encrypts GitHub tokens and webhook secrets with a generated key and stores review history in a local SQLA database. I showed how to run ReviewLogicFlow, post the agent comment to GitHub, view analysis history, and check last reviewed results. There was no specific action requested from viewers."}