What we store, and why.

The short version

We store the email address you sign up with, the brands and prompts you configure, and the answers the AI engines return when we run those prompts. We do not sell any of it, we do not use it to train anything, and we do not run advertising trackers. You can ask for all of it to be deleted and we will do it.

What we collect

Your account

An email address, a display name if you give one, and either a password hash or the fact that you sign in with a link or with Google. If you sign in with Google we receive your email address and name from Google; we do not receive or store your Google password.

Your workspace

The brands, competitors, prompts, engine settings and schedule you configure, plus the members of your workspace.

Provider API keys

If you bring your own keys, they are encrypted at rest with AES-256-GCM before they touch the database. The application returns only the last four characters, and we have no feature anywhere that displays a full key. They are decrypted in memory for the duration of a single call to that provider and never written to logs.

Run data

For every scheduled or manual run we store the question asked, the full answer the engine returned, the sources it cited, the searches it ran, the model used and what the call cost. This is the product: the metrics are derived from these answers at read time, which is what lets a metric definition change without orphaning your history.

Operational data

Ordinary server logs, including IP addresses, for security and debugging. Vercel Analytics on the marketing site and dashboard, which is cookieless and does not track individuals across sites.

What we do not do

• We do not sell or rent your data to anyone.
• We do not use your prompts, answers or workspace content to train any model.
• We do not read your workspace content except when you ask us to help with a problem.
• We run no advertising or cross-site tracking.

Who else sees it

Running a prompt means sending it to the AI provider you selected, so the text of your prompts reaches OpenAI, Anthropic, Google, Perplexity or OpenRouter depending on which engines and routes you enable. Their terms govern what they do with it. If you bring your own keys, that traffic is on your account with them.

For engines without a public API, and for the consumer-app route of ChatGPT, Gemini and Perplexity, the question text goes to the third-party provider whose key you add, or ours on Pro.

The other processors we use to run the service:

• Neon, the managed PostgreSQL database.
• Vercel, which serves the marketing site and the dashboard.
• Resend, which delivers sign-in, verification and invitation emails.
• Stripe, which processes payments. Payment details go to Stripe and never reach our servers. We store a Stripe customer identifier and your subscription status.
• A virtual server that runs the API and the worker.

How long we keep it

Run data is kept until you delete the brand it belongs to or your workspace. Deleting a workspace removes its stored API keys, brands, prompts and every collected answer. Backups are retained for up to 30 days, so deleted data can persist in a backup for that long before it ages out.

Getting your data out, or deleted

Every screen has a CSV export, which is the fastest way to take your data with you. For deletion, or for a copy of anything the export does not cover, email [email protected] from the address on the account and we will action it within 30 days.

Depending on where you live you may have rights to access, correct, export or erase your personal data, and to object to processing. The address above is how to exercise any of them; we do not require a particular form of words.

Security

Provider keys are encrypted at rest. Traffic is HTTPS end to end. Workspaces are isolated from each other and that isolation is covered by automated tests. Security reports are welcome: email [email protected] with "security" in the subject line and we reply before we act.

We are a small project and will not pretend otherwise: there is no SOC 2 report and no 24/7 security team. If that matters for your use case, tell us what you need and we will say plainly whether we can meet it today.

Children

CitePrism is a business tool and is not directed at anyone under 16.

Changes

If this policy changes in a way that affects what we do with your data, we will email account holders before it takes effect rather than quietly changing the date at the top.

Contact

[email protected] for anything on this page.