<?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/a8724281fef94ab5817eb2fde60be60b&quot; frameborder=&quot;0&quot; width=&quot;1914&quot; height=&quot;1435&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>1435</height><width>1914</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>1435</thumbnail_height><thumbnail_width>1914</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/a8724281fef94ab5817eb2fde60be60b-a6a6f7c60bb380d4.gif</thumbnail_url><duration>252.697</duration><title>Automating Image Prompts with n8n and Pickaxe</title><description>This Loom discusses how to automate sending an uploaded image from Google Drive into a GPT “pickaxe” workflow to generate a microblog. The author shows an existing method that works by uploading a photo directly and asking for a microblog, then proposes triggering the same behavior via n8n when a new image is added to a Drive folder. They tried testing by sending a link but received an error that it cannot access Google Drive, and they are looking for the right n8n action to let the workflow fetch the image. They also mention using GPT 5.5 image capabilities and wonder whether they should use a custom HTTP request or handle binary data, and ask if the community node has been updated since it was created.</description></oembed>