Oathscript is the commitment-tracking layer for customer support teams who already know response-time SLAs miss the point. The reply is the easy part. The promise is the hard part — and the part that keeps the customer.
HubSpot, Gorgias or Zendesk — $99 / month to 2,000 tickets, then linear.

“Refund the $189 annual add-on by Tuesday, and add three months of credit for the outage.”
A response is a reaction. A promise is a commitment.
Every support agent — and every AI assistant next to them — makes specific promises every day. A replacement by Friday. An escalation within forty-eight hours. Three months of credit. And most of those promises vanish into the transcript the moment the ticket auto-closes.
Dashboards that count replies do not see this. CX leaders do. Oathscript is built for the people who feel the gap between “we replied in eleven minutes” and “we are still two weeks late on the refund” — and want the second number to be visible, trackable, and owned.
No dashboard migrations. No retraining. Drop in an export and the first ledger is ready before your standup.
Drop in your HubSpot or Gorgias ticket export. Daily, weekly, or on-demand — Oathscript reads transcripts and proprietary fields, never touches production.
Every natural-language commitment is lifted out with its customer, owning agent, due date, financial value, and ticket link — not a fuzzy embedding, a record.
Nothing is created or closed without an agent ’s sign-off. That single rule is why precision stays high and the noise of naive RAG pipelines stays out.
Daily briefs list every open and overdue promise. The dashboard ranks who is holding and who is slipping — by agent, by team, by week.
Your dashboards count responses. Your customers remember commitments. Oathscript closes that gap — every "refund by tomorrow" and "we’ll replace this on Friday" becomes a record with a due date and an owner.
A promise is born only when the agent responsible confirms it. A promise is fulfilled only when the agent confirms that, too. The model proposes; the human disposes. False positives stop at the door.
Pavel has 3 candidate promises waiting on a confirm. Two will approve, one will reject — the AI was too generous with “we’ll look into it.”
Every morning, every agent and every manager gets a tight email: who owes what to which customer, what is overdue, what is disputed. Read in sixty seconds, act on in five.
Beyond the dashboard, Oathscript is building the dataset that does not yet exist anywhere: structured promise patterns linked to refunds, escalations, and cancellations. The view of CX that no BI tool gives you today.
CX operations leaders
You measure promise-to-fulfillment and refuse to ship a dashboard that is just a count of replies.
Support managers
You need every agent’s commitments visible on Monday morning — not reconstructed from a Slack scrollback.
Trust & safety teams
You already track escalations; you need the structured set of promises that preceded them.
HubSpot, Gorgias, Zendesk shops
You have the transcripts. You have the volume. You are tired of looking at response-time dashboards.
Predictable math so growth does not buy a surprise invoice. No per-seat games, no enterprise-unlock-gates for things you actually use.
Up to 2,000 tickets processed monthly. Built for a single support team trying Oathscript on for size.
Scaling
Linear volume tiers above 2k.
Beyond 2,000 tickets, pricing scales line-by-line so growth does not buy a surprise. Bring the whole CX org.
Scaling
Volume scales the same way.
For trust-and-safety, escalations, and regulated CX. SAML, custom retention, and the outcomes dataset add-on.
Scaling
Annual, with procurement-ready paperwork.
If yours is not here, write to us and we will answer in plain language.
We are onboarding the first cohort of CX teams. Tell us what you promise, what slips, and what you would like to see next.
We read every message at oathscript@polsia.app.