Moda introduces editable designs from 1 prompt with PNG and PDF export
Moda says a single prompt can generate editable designs and presentations on a real canvas, with redesign-from-reference support and export to PNG or PDF. The release targets users who need assets they can still tweak after generation instead of a flat image.
TL;DR
- Moda says a single prompt can generate editable designs instead of a flat image, and AmirMushich's demo clip shows the canvas staying editable after generation.
- In AmirMushich's workflow post, the product flow is prompt, generate, edit, then export, with PNG and PDF called out explicitly.
- Moda's official homepage positions the tool around slides, social posts, docs, carousels, and beta website or HTML document outputs, all on a "real canvas" users control.
- Moda also says users can redesign reference images, and AmirMushich's thread frames that as especially useful for non-designers.
You can open Moda's homepage and see pitch decks, QBRs, social posts, and employee docs in the example gallery. The linked MCP docs go further than the tweet, showing Moda wired into Claude workflows, with support for references like PDFs and existing Moda files. There's also a setup page for Moda MCP that lists Claude, Codex, and Cursor as clients.
Editable canvas
Moda's core pitch is simple: generate a design from text, then keep editing it.
The official homepage makes the same claim in product language, "Finally, AI designs you can edit," and pairs it with a broader asset menu than the tweet alone shows: slides, social posts, docs, carousels, plus beta website and HTML document formats.
Prompt-to-design flow
The workflow in AmirMushich's post breaks into four steps:
- The agent reads the prompt.
- It designs the asset.
- You edit the result.
- You export it as PNG, PDF, and more.
That matches the sequence in Moda's Claude design docs, where Claude starts a design task, polls until it is ready, then returns a Moda canvas URL for further edits or export.
Reference-driven redesigns
Moda is not limited to greenfield generation. In AmirMushich's thread, the product is also pitched as a redesign tool for reference images.
The more detailed MCP documentation expands that reference idea beyond images. It says users can hand Claude reference images, PDFs, or existing Moda designs, then turn those inputs into decks, posts, ads, or reports.
Access and agent support
Moda's homepage ends with "Get started for free," which is the clearest availability signal on the public site.
A separate Moda MCP page shows where the product is already being threaded into agent workflows. It lists Claude.ai, Claude Desktop, Claude Code, Codex, and Cursor, and says setup runs through Moda's MCP server plus an installable skill.