Your content stays with you.

In the basic generator, website addresses, Wi-Fi names and passwords, uploaded logos, and card text are processed in your browser. They are not sent to our server or stored in browser storage. Reloading the page clears your work.

Google accounts and payments

Agent requires Google sign-in. We store your Google account identifier, verified email and a revocable session. We do not receive your Google password or access your Gmail messages. An essential secure cookie keeps you signed in for up to 30 days. Sign out to revoke it.

Daily grants, credit reservations, usage entries and payment references are stored to operate your wallet and reconcile payments. Stripe processes card details; our application does not store them. Account, payment and Agent pages are excluded from session recording.

Paid Agent results are retained for up to one hour for retry delivery, then removed automatically. Task inputs are not stored by our wallet. Credit and order records remain for account operation and payment reconciliation.

Simple, content-free analytics

We count page views, starts, successful exports, and validation or export failures, together with the selected content type and file format. Failure events do not include error messages. These events contain no QR content, uploaded files, query strings, cookies, or persistent user identifiers. We honor your browser’s Do Not Track setting.

When hosted, Cloudflare handles network requests and may process connection information such as IP addresses for delivery and security. Our application does not include IP addresses in analytics events.

Microsoft Clarity

On our production website, Microsoft Clarity loads only after you allow analytics. It helps us understand page usage through heatmaps and session recordings. The editor and QR previews are masked, and advertising consent stays off. Local and preview sites do not record sessions.

Use Analytics settings to turn it off at any time. This stops further recording in this page without clearing your QR work; it does not delete recordings already sent. If you allow it again, recording resumes on the next page load. We store your choice for up to 180 days and honor Do Not Track and Global Privacy Control. See Microsoft’s Privacy Statement.

Art, print, and batch workspaces

Local artwork, reference images, CSV generation, and print exports run in your browser. Moving a design between workspaces keeps it temporarily in memory; we do not put the content in URLs or browser storage. Reloading or document navigation clears it. Sensitive art, print, batch, and Agent workspaces are excluded from Clarity recording.

Agent AI features

When enabled, AI planning sends the request you explicitly approve to Workers AI. Batch planning sends only your instruction, headings, row count, and current plan, never CSV cells. Planning also includes built-in product guidance maintained by Borealbit. Uploaded files are never used as knowledge sources. Agent revisions include the current reviewed plan. Art generation sends only your description; the QR link and reference image are not included. Requests are rate limited using network information and your signed-in account. A successful paid result is retained for retry delivery for one hour. Flue stores the model conversation and skill steps in a private Durable Object for up to one hour. Your approved request, resulting plan or image prompt, model usage, and prompt version are also saved as account history until you delete them. Image bytes are stored in your library only when you save the PNG. The local tools remain available without an account or AI.

Saved plans and files

Cloud saving is explicit. Save buttons upload the selected CSV, artwork PNG, or reviewed plan to your account library. Cloudflare R2 stores original files privately; D1 stores ownership, file metadata and saved plans. Saved items remain until you delete them in Saved projects & files. Unsaved browser work remains temporary. CSV cells and uploaded images are not indexed as knowledge. Product knowledge is built in and maintained by Borealbit; customers cannot upload or modify it.

Background exports

Starting a background export explicitly uploads your CSV and reviewed plan to private R2 storage. A Cloudflare Workflow creates the ZIP; D1 stores its owner and status. CSV cells are never sent to a model. Up to five exports, each with up to 50 rows and a 20 MB output, stay available until you delete them. The normal Generate batch button continues to work locally.

Private Agent integration

The basic generator above stays local. The separate Agent integration processes a CSV on our server only when its caller explicitly requests that processing. Model-assisted planning also requires a separate choice: the instruction, column headings, row count, and previous plan go to Workers AI. CSV cell values are not included in the model request. Do not include passwords or private data in instructions or headings. Agent files are returned to the caller. Server tasks keep only their operation, status, progress counts and timestamps for up to one hour. Instructions, URLs, CSV contents and generated files are not saved in task storage. Downloads can contain private URLs.

Your downloads

A downloaded Wi-Fi code contains the network credentials. Share or print it only where you want to grant access. A static website code links directly to the address you entered; it cannot be changed after printing.