FAQ

The Convas Agent is built for presentation work. You describe what you need in chat; it organizes sources, outline, and design in one project, then renders slides page by page—with web search, image generation and editing, document intake, and more. Compared with a general coding agent figuring everything out from scratch, Convas uses a deck-specific toolchain and render pipeline, so a full deck usually uses fewer tokens. See the Agent overview doc for details.

Convas uses a chat Agent to build and refine decks across four tabs, with click-to-edit on the canvas, share links, and multi-format export.

No. Brief, outline, design, and canvas are always available—the Agent works on one deck throughout, with no fixed order.

It depends on the task. The current chat turn—streaming reply, parsing documents, drafting an outline—keeps running on the server after you close or refresh, and you see the saved result when you reopen; if you reopen or switch tabs mid-turn, the in-progress output replays from the start and keeps streaming. To stop the turn, use the ■ button beside the composer. Batch slide rendering continues on the server: chat may finish quickly while the canvas tray and composer show background rendering or queued status; reopen the project to see progress. Switching Brief, Outline, Design, or Canvas inside the workspace is usually fine.

Code mode produces editable pages with canvas refine and Tweak; image mode produces full-page images, with revisions mostly via chat regeneration. You choose when creating the project. See Output modes for details.

In chat, name the page number or select an element on the canvas, e.g. “Change the title on slide 3.” The Agent updates just that page.

Six ratios: 16:9 widescreen, 4:3 classic, 1:1 square, 3:4 portrait (social cards), 9:16 vertical, and 21:9 ultrawide. You choose when creating a project; the workspace header shows it as a badge.

Share for online viewing; export for offline meetings, archives, or PowerPoint edits. See Share and Export docs.

No. Export freezes motion to a static frame; browser preview and share links still play motion. See Motion & animation.

Image generation produces bitmaps (PNG, etc.)—best for photo-like mood art. SVG generation creates scalable vector illustrations from text; vectorization turns an existing bitmap into SVG—best for simple icons. Flow and architecture diagrams are drawn directly at render time instead. See Visual elements.

Models and actions consume Credits; Free and paid plans include different allowances. See the pricing page for details.

Your deck lives under your account. When you use AI features, content you submit is sent to third-party AI providers to complete generation—we do not use your projects for AI model training. Share links show a read-only view—visitors cannot edit your project.

Try clearer prompts, make sure your brief is complete, or ask the Agent to plan first. If issues persist, email support@convas.design.

Email support@convas.design. Refunds and acceptable use are in the Terms of Service; personal data is covered in the Privacy Policy. To delete your account, write from your registered email—we typically complete this within 30 days, except records we must keep by law.