Privacy Policy

Gaurav Saxena operates the VivahVerse ChatGPT biodata integration. When you request a preview, we receive only the supported details you confirm and choose to send, such as your name, birth date/time, education, profession, family details, contact details, and an optional separately selected portrait. We use these details only to compose the requested biodata and perform an optional astrology calculation you request. We do not use this integration's biodata for marketing, advertising, matching, profiling or model training.

ChatGPT reads any biodata, PDF, image or Kundali you upload to the conversation. VivahVerse does not receive those original source documents through this integration; only confirmed supported values are sent. A portrait must be selected separately and must not be an entire biodata scan. Do not submit passwords, government identifiers, payment-card details or protected health information. Raw street addresses and raw birth-location text are not supported in the model-facing tool schema. Religion, caste, gotra, complexion and free-form sections remain disabled in this release. Choosing a design category does not disclose or establish your religion.

Astrology is optional. We reuse values you confirm from an existing Kundali or biodata unless you ask to recalculate. If you request calculation, you directly select a birth-city reference in the calculator and confirm the birth date and local time. These values are processed in memory by the same local panchangam-js engine used by our website. The plugin uses a bundled GeoNames city catalog and historical timezone rules, so it does not send your birth details to an external geocoding API. City centroids and traditional astrology outputs are estimates; review them before inclusion. We do not infer caste, health status or compatibility scores from the calculation.

Preview values and selected portraits are processed transiently in a fresh isolated browser context in Vercel rendering compute. The context closes after the operation; short-lived rendering environments stop automatically. Customer details are not included in reusable browser snapshots. Previews and calculated values are returned to ChatGPT, which may retain the conversation and tool results under its own policies. Our website, MCP endpoint and public template assets are hosted by Vercel. Temporary generated PDFs and cross-instance operational state use private PostgreSQL storage hosted by Supabase. A legacy private test connection may still use our separately operated VPS until it is retired; new Next.js connections use Vercel rendering compute. The service uses OpenAI/ChatGPT file delivery only for an explicitly selected portrait on an approved host. No original source documents are forwarded to these providers by the biodata tools, and no email delivery, advertising SDK or account registration is used by this integration. The private MCP schema stores encrypted temporary PDF payloads, hashed request/capability identifiers, short-lived rate-limit records and authenticated render-worker metadata; it is not exposed through the public database API.

Generated PDFs in the Next.js integration are encrypted before storage in a bounded private database store. Links stop working after 15 minutes according to the database clock, even before cleanup. A scheduled job runs every minute to remove expired active records. The legacy VPS connection uses a bounded memory store and clears expired bytes within a further 10 seconds. You can request earlier deletion in ChatGPT using the PDF's private deletion capability. Deletion removes the active stored record; provider backups and infrastructure records have separate retention and are not immediately erased by this operation. Restarting Next.js compute does not remove shared database records. Restarting the legacy VPS service removes its memory-only PDFs. Download links are not account-bound: anyone holding the link can access the PDF before expiry or deletion. We cannot remove copies already downloaded, retained by ChatGPT or shared by you; manage those copies in the corresponding service.

For service reliability we record only non-content operational events: a random request identifier, result code, duration and fixed error code. We do not intentionally log biodata values, portraits, original files, IP addresses, download/deletion capabilities or complete request URLs in application logs. Vercel and Supabase manage infrastructure and non-content log retention under their provider policies; we do not promise immediate removal of provider logs or backups. Legacy VPS application logs rotate at 10 MiB per file, with three files retained. Rate limits use keyed hashed client identifiers in bounded shared database records for approximately one minute in Next.js, or bounded process memory in the legacy service. Hosting/network providers may maintain their own infrastructure records under their policies. This integration does not use browser analytics or advertising trackers.

You may leave optional details out, correct values before preview, choose not to calculate astrology, or ask to delete a temporary PDF. Requests concerning VivahVerse processing can be sent to support@vivahverse.in. Requests to erase a ChatGPT conversation or an independently downloaded file must also be made in the relevant host or device. Supply details about another person only with their permission. This document describes service behavior and does not promise that every uploaded document is readable or every estimate is accurate.

When you join our early-access list, we collect your name and email address. This information is provided voluntarily and is used solely to communicate product updates and launch information.

When you use a free reflection tool, your answers stay in your browser unless a feature clearly says otherwise. If you ask us to email a reflection, we store your email address, the requested delivery, and any optional communication preference you choose. The attached PDF is held temporarily to deliver your request, then deleted after the email provider accepts it or within 24 hours. We do not create a separate record of your reflection answers. Delivery contact and metadata are retained under the tool’s existing retention policy.

We use the information you provide to send relevant product updates, early-access invitations, and occasional announcements about Vivahverse.

A reflection email is transactional and is sent only when you request it. Practical guidance is sent only when you separately opt in. You can unsubscribe from practical guidance without affecting access to your reflection email.

We do not sell, rent, or share your personal information with third parties for marketing purposes. Your data is never used for advertising.

We use HTTPS for transport and access controls appropriate to each feature. Storage and processing differ by feature as described in this policy; no online service can guarantee absolute security.

We retain tool email contacts until you unsubscribe or request deletion. We retain delivery metadata for up to 90 days for support and duplicate-prevention purposes. You can request deletion of your data at any time by contacting us.

We use minimal, privacy-respecting analytics to understand how visitors interact with our website. We do not use invasive tracking cookies or third-party advertising trackers.

Essential cookies may be used to maintain your theme preferences and session state. These do not track your activity across other websites.

You have the right to access, correct, or delete your personal data at any time. You can unsubscribe from our communications using the link provided in every email.

If you have any concerns about how your data is handled, please contact us at hello@vivahverse.in and we will respond within 48 hours.

We may update this privacy policy from time to time to reflect changes in our practices or for legal, regulatory, or operational reasons. We will notify you of any material changes via email or through our website.

This policy was last updated on 24 October 2026.