How we handle your data. Short version: we log what's needed to run the product and nothing more.
We collect three kinds of data:
We do not store the content of your prompt bodies or model outputs for BYOK traffic. We retain the rendered prompt briefly during execution, then drop it.
We use your data to:
We do not sell your data. We do not use your prompts or outputs to train our own models. We never will.
We retain each class of data only as long as we need it:
Deletion is automated via a daily retention sweep. Anonymized aggregate statistics (e.g. "prompts run this month") may persist indefinitely — they contain no personal data.
Under GDPR, CCPA, and similar frameworks, you have the right to:
You can exercise access, export, and deletion rights directly from Settings → Privacy & data — no email required. For correction, objection, or anything else, email privacy@oort.dev. We respond within 30 days.
When you run a prompt or workflow, your request is forwarded to the model provider you selected (Anthropic, OpenAI, etc.). That provider processes the request under its own privacy policy, using your API key. We recommend reading the privacy policy of whichever provider you route to.
We use the following infrastructure partners: Supabase (Postgres, auth), Cloudflare (CDN, DDoS), Stripe (payments), Sentry (error monitoring). Each is bound by a data processing agreement.
Oort is not for anyone under 18. If we discover a minor has created an account, we'll delete it promptly.
We'll post material changes at least 30 days before they take effect, via email and in-product notification.
Privacy questions, data requests, and complaints go to privacy@oort.dev. EU residents also have the right to lodge a complaint with their local data-protection authority.