{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/3bba72893bb846ecab4279c597068d73\" frameborder=\"0\" width=\"1660\" height=\"1245\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1245,"width":1660,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1245,"thumbnail_width":1660,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/3bba72893bb846ecab4279c597068d73-ba9d361a8d4c84f3.gif","duration":540.624,"title":"Building Apps with Claud Code and Codex","description":"This Loom walkthrough focuses on the Farm to People app that uses browser automation and AI to suggest meals and add-ons. The author scrapes their Farm to People cart via Playwright, then uses Haiku for meal titles and Sonnet for recipe types, streaming updates so users see progress quickly. They explain pulling quick add-on data from the site, generating pro tips and rationale for how add-ons complement meals, and offering chat prompts like healthier meal requests. They also briefly cover Sandlot, including twice-daily scraping and an AI assistant using DeepSeek V4 Flash, and mention an approval-driven scheduled system using a Zocomputer-based setup."}