Privacy policy
Updated October 29, 2026
This policy explains how Refcat handles personal data in the website, application and extension.
1. Data
Google sign-in supplies your name, email and profile image. We store account preferences, saved references, projects, export history and the content you explicitly submit for AI processing. Browser export authorization records the page title and URL in your account's history.
2. Purposes
We use data for authentication, requested features, payment, security, reliability and service communication. We do not sell personal data.
3. Providers
Depending on the enabled feature, providers include Google (authentication), Stripe (payments), Amazon SES (email), OpenRouter and Anthropic (AI), Cloudflare (website hosting, CDN and object storage) and configured server hosting. Prompts and selected images are sent to the AI provider when you invoke the feature; Anthropic processes project chat, Markdown exports and design system generation. Configured Chatwoot handles messages you send through support chat; its widget loads when you open support. Himetrica (product analytics and error monitoring) and configured OpenReplay (session replay and live support co-browsing) receive data only under the analytics choice below. The public website uses Google Analytics, Google Ads measurement and Himetrica to measure visits only after you accept its cookie banner; until then Google tags run with every consent signal denied and Himetrica does not load. Do Not Track and Global Privacy Control count as a refusal.
4. Analytics
Optional analytics in the application is off until you allow it in the banner shown after sign-in or in account settings on this device. Do Not Track and Global Privacy Control turn it off. On the public website, the cookie banner controls the same first-party measurement: with your consent, each page view and the start of each visit (page route, language, theme, a random visit ID and an anonymous browser ID, kept in local storage as refcat:anon_id and refcat:product_session) are sent to Refcat's API at api.refcat.app. The request carries your Refcat session cookie, so if you are signed in on that browser these events, and earlier anonymous events from it, are linked to your account. Visit starts also go to Himetrica and Google Analytics. Without consent nothing is sent, and declining removes both identifiers. With your consent: usage events go to Refcat's database (prompts, raw errors and captured URLs are excluded); Himetrica receives page views, product events, uncaught errors, browser console messages and performance metrics, linked to your account, name and email; and, when configured, OpenReplay records catalog page structure and interactions with typed input, code blocks, dialogs and the sidebar hidden, network and iframe capture disabled and URL parameters removed. Authentication, billing, backoffice, private projects/folders and searches are excluded from replay. When you open support chat, replay covers every page except the backoffice, shows dialogs and the sidebar and is linked to your email so support can find it; a support agent can watch your screen live only after you approve an on-screen request, and screen control or a call asks again. Declining or turning analytics off stops all of this. Product operations also record server events for quotas and reliability. The automatic extension screenshots described below are a separate feature.
5. Retention
Product event records expire after 400 days. Account data is kept while needed to provide the service. Financial records and email suppression choices may be retained for billing, security or legal obligations. Provider retention is governed by the applicable service agreement.
6. Choices and security
You can edit your profile, disable optional analytics, unsubscribe from marketing email and request account deletion. Privacy requests cover retained data and configured providers. We use technical safeguards, but no online service can guarantee absolute security.
7. Updates
We publish policy changes here and update the date. Significant changes can also be communicated in the application or by email.
8. Automatic private screenshots
Extension data collection is optional and off by default. Nothing is collected until you turn it on in the extension preferences, shown at install; you can turn it off there at any time, before or after sign-in. Preferences saved while collection was pre-selected by default are treated as off. Existing disabled settings and site exclusions remain respected. Human review is optional and requires a separate choice. When enabled, each opening sends one visible screenshot, site origin, account ID and name, date, policy version and, only when human review is enabled, recorded authorization time. Each completed Figma export also sends its .fig file, linked to that export's record. While the extension is open on a page, clicks on that page send their position, a short element path (tag, id, up to two classes), viewport size, the page origin and path, and elapsed time; typed text, element text, query parameters and fragments are not sent. Screenshots and .fig files may contain personal or client data; share only with permission. Excluded sites and subdomains are not collected. Screenshots and .fig files are encrypted, excluded from public catalog and search, and not sent to AI providers. Authorized reviewer access is recorded. Delete individual screenshots in Account > Exports. Account deletion revokes access immediately, deletes click records and queues object cleanup with retry. Screenshot retention is 90 days; operators remove expired objects and records, with a seven-day handling target. Disabling collection does not delete data already collected. Figma export works independently.
9. Controller and contact
Refcat is the controller of the personal data described here. Send privacy requests, and questions for the person responsible for data protection (the encarregado under Brazil's LGPD), to hello@refcat.app.
10. Legal bases
We process account, sign-in, requested features, payments and service emails to perform our contract with you (GDPR art. 6(1)(b); LGPD art. 7, V). Security, abuse prevention, rate limits, reliability and the server events behind quotas rely on our legitimate interest in running a safe service (GDPR art. 6(1)(f); LGPD art. 7, IX). Optional analytics on the website and in the application, session replay and extension data collection rely on your consent (GDPR art. 6(1)(a); LGPD art. 7, I), which you can withdraw at any time without affecting earlier processing. Financial records kept for tax and accounting follow legal obligations (GDPR art. 6(1)(c); LGPD art. 7, II).
11. International transfers
The providers listed above may process personal data outside your country, including in the United States. Write to hello@refcat.app for information about the safeguards that apply to these transfers.
12. Your rights and complaints
You can ask for confirmation that we process your data, access, correction, portability, anonymization, blocking or deletion of unnecessary data, information about who we share it with, and you can object to processing or withdraw consent, by writing to hello@refcat.app. You can also complain to Brazil's National Data Protection Authority (ANPD, gov.br/anpd) or, in the European Economic Area or the United Kingdom, to the supervisory authority where you live, work or where the issue happened.
13. Microsoft Clarity
Microsoft Clarity receives page structure and interactions on the public website and dashboard to generate heatmaps and session replays, only after you allow analytics. The same choice and browser opt-out signals control it. Input values are masked by default. We do not explicitly send account IDs, names or emails to Clarity. Turning analytics off stops collection.
For access, correction, deletion or privacy questions: hello@refcat.app.