Talk to a real human.

Proxima is a small team of AI agents that read your email, watch your calendar, and find your leads — quietly, in your team Discord. Every outbound action passes through you for approval. Read the docs for the longer answer.

No. Every outbound action — every email, every reschedule, every DM — surfaces in Discord as an approval card. Approve, edit, or skip. The approval contract is the whole point.

Open workspace Settings → Danger zone → Delete workspace. We delete the workspace record and every credential within 7 days, every message and audit log within 30 days. For per-user personal data, see Privacy → Your rights.

No. Your data is used to run the product — generating drafts and reasoning about your work — and that’s it. We do not train any model on your data, and our LLM provider (Anthropic) does not train on data sent through their API. See Privacy → Model training.

The corresponding agent stops working immediately and the OAuth refresh token is revoked on both sides. Drafts that were waiting on approval stay in your audit log; nothing already approved is undone.

Yes. Email hello.proxima@dterminal.net from the workspace owner’s account and we’ll prepare a JSON export of every workspace credential, agent context, and audit-log entry. This is a manual process today; an in-product export is on the roadmap.

Please email security.proxima@dterminal.net with details. We respond within one business day and don’t pursue legal action against good-faith reporters. See Security → Disclosure.

Currently in Google Cloud, US-based regions. We’ll add EU co-location once we have demand for it. See Security → Infrastructure for the full breakdown.

Free during early access. Paid plans land later this year, with at least 60 days’ notice before any change to early-access workspaces. We’ll never charge silently.