What we collect
From the submission form:
- Name and email. Required so we can write back to you about whether Argus can check your question, agree its scope, and send you the session.
- The question itself: the one-line question, what a wrong answer would look like, where the data lives, and any other notes you provide. We treat these as research material that is published as part of the resulting session if we run it as public research.
From your interaction with the site:
- Operational traffic data. Our proxy and application may process IP address, timestamp, request path, and user agent to deliver the site, investigate failures, prevent abuse, and enforce rate limits. Operational logs are size-rotated and access-restricted; their lifetime varies with traffic.
- Self-hosted, cookie-free analytics. We use a self-hosted Umami instance for aggregate page views, referrers, browser/operating-system/device categories, approximate country, and named product events such as opening evidence or submitting a request. We do not use Google Analytics, advertising pixels, or session replay, and we do not use Umami to identify a submitter. The tracker honours your browser's Do Not Track preference and strips query strings and URL fragments before collection.
- Privacy-bounded interaction analytics. Our first-party collector keeps cumulative per-page counts for clicks on stable interface identifiers, scroll reach, active time, sections viewed, form fields touched, validation failures, outbound-link use, coarse viewport size, JavaScript errors, and Core Web Vitals. It also estimates repeated or ineffective clicks so we can find confusing interfaces. These summaries are sent about every 15 seconds and when a page is left. We do not collect typed values, individual keys, selected text, raw pointer coordinates, DOM snapshots, or a replayable recording. We do not fingerprint your device. Query strings and URL fragments are not included, and Do Not Track disables this collector.
- First-party browser storage. Argus stores a random identifier and your per-session helpful/not-helpful choice in local storage. A separate random journey identifier lives only in tab-scoped session storage. The server stores only a one-way salted hash of the browser identifier. This supports deduplicated feedback and aggregate reader counts without storing the raw identifier. Neither identifier is joined to your name or email or used across other sites. During a submission, tab-scoped session storage also keeps a random retry key indexed by a SHA-256 fingerprint of the form payload. The form contents themselves are not stored there; the key prevents a lost response and reload from creating a duplicate. Some published sessions also remember a theme choice in local storage. The public site sets no advertising or analytics cookie.
- Fonts. The site serves its fonts from its own server and makes no third-party font request.
What we do with it
We use what you submit to decide whether Argus can check your question, to agree its scope with you, to run the session, and to send it to you. We use server logs to keep the site secure and stable.
We do not sell or rent any personal data. We do not use submissions to train any model owned by Obvix Labs or by any third party. We do not share submissions with marketing partners. We do share information with infrastructure providers (email delivery, web hosting, and the language-model or retrieval APIs used during an accepted session) only to the extent required to operate the service and under the provider terms applicable to our account.
A public question we run is published with its whole session, and you choose how you are credited: by name or anonymously. Your email and IP address are never published, and your notes are not quoted unless you ask.
Where data lives
Submission data is stored on infrastructure operated by Obvix Labs and on services we use to run Argus (transactional email delivery, model APIs, and so on). Our public analytics database is self-hosted with the application. Some processing may take place outside India where a provider is based.
We are aligning this service with India's Digital Personal Data Protection Act, 2023 and the Digital Personal Data Protection Rules, 2025 as their provisions take effect. Where GDPR or UK GDPR applies, the legal basis depends on the activity: responding to your request, operating and securing the service, or consent where we specifically ask for it. You can ask us for the basis applying to your record.
Retention
Published sessions and their revision history are retained as the public record. Contact details are not part of the public artifact and can be corrected, erased, or separated from that record where the law permits.
Questions we do not run are kept for 24 months in case we can take them on later, then deleted. We tell you by email when we decline a question, but we don't guarantee deletion timing more granular than this.
Raw self-hosted analytics is kept for up to 24 months and then deleted or reduced to non-identifying aggregates. Size-rotated operational logs are overwritten as their configured bounds are reached. Email and rights-request correspondence is retained only as long as needed for the request, security, dispute, or a legal record-keeping obligation.
Your rights
You can write to us at any time at connect@obvix.io to:
- Access the personal data we hold about you.
- Correct any personal data that is inaccurate.
- Withdraw a question at any time before we begin running it. Once a session is published, you may ask for the credit to be made anonymous.
- Delete your personal data, subject to the retention requirements in §4 and to any audit-trail requirements that apply to a published session already credited to you (we can make the credit anonymous; we cannot remove the published session itself).
- Object to a particular use, including the use of quoted portions of your submission in a published session, before publication.
- Lodge a complaint with your local data protection authority if you believe our processing is unlawful. For India, that is the Data Protection Board of India; for the EU, it is your Member State's authority; for the UK, it is the Information Commissioner's Office.
We aim to respond to rights requests within 30 days. We will verify your identity before acting on a request that would affect personal data; this is to protect you from third-party impersonation.
Children
Argus is not directed at people under 18. We don't knowingly accept submissions from minors. If you believe a minor has submitted a question, write to connect@obvix.io and we will remove it.
Security
We take reasonable technical and organisational measures to protect submission data: TLS in transit, separate operator and public-ingress credentials, deny-by-default private access, restricted credential files, bounded logs, and checksummed backups. At-rest protection also depends on the production host and backup storage configuration; we do not claim that application-level controls make every copy risk-free. If we become aware of a breach affecting your personal data, we will make notifications required by applicable law.
Changes to this policy
We may update this policy as Argus evolves. Material changes will be announced on the site and (for users with an active submission) by email. The effective date at the top of the page always reflects the current version.
Contact
Questions, requests, or complaints about how we handle data: connect@obvix.io. We treat this as both the privacy contact and the grievance officer contact for purposes of the Indian DPDP Act and analogous regimes.
Last updated: 14 November 2026.