Frequently Asked Questions
Quick answers about transformations, Agentic Transform, and publishing.
Automations run AI and media services on your project files. You can use focused steps
(text-to-image, text overlay, caption generation, video) or Agentic Transform,
where an agent chooses tools and retries after a vision readability check until it has
a usable graphic.
You can transform text/documents, images, audio, and video. Common paths include captions from
images, images from text, text overlays on graphics, agentic tip graphics from documents,
merge-layers compose, and video/audio generation.
Agentic Transform is a document automation that follows your instructions and uses tools
(text-to-image, text overlay, compose) to produce a social graphic. After compose, vision QA
checks that the overlay text is readable. If it fails, the agent adjusts size, contrast, or
copy and tries again. Activity logs show each step and any rejection reasons.
Yes. Apply the Website Service Ads workflow preset and paste your site URL.
We scrape the site, import your offerings, benefits, CTA, logo, and brand styles, then generate
square Instagram ads that promote what your business does. Connect posting when the outputs look right.
Single-step transforms depend on your prompts. Agentic Transform adds an automatic vision
readability check for on-image text so weak overlays are retried instead of silently shipping.
Always preview important posts before publishing.