{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/284c0753a76f48ecaa2d7c16757928d4\" frameborder=\"0\" width=\"1898\" height=\"1423\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1423,"width":1898,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1423,"thumbnail_width":1898,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/284c0753a76f48ecaa2d7c16757928d4-1fe7d46a7b98c77f.gif","duration":76.005,"title":"Auto-Updating WAPI via Sakura Evaluator Workflow 🚀","description":"Hi, we built an app that uses Sakura's evaluator platform and connects to a WAPI agent. When you run the Sakura evaluator and get a result, our agent retrieves it and can auto update WAPI configuration. In our simple evaluator example, the agent must say hello first. I show how to start a new evaluation, then when it is running, it updates the system prompt and you rerun the evaluation so from round two you can see results. No direct viewer action was requested in the video."}