Runs on Google Calendar, MongoDB Atlas, OpenAI, Vercel, AWS and Resend
Point it at what you already have
From your own documents, with the source shown under every reply — and an honest “I don't know” when your files don't cover it.
It reads your Google Calendar free/busy, holds the slot, and emails the confirmation to both of you. Not a link — the booking.
Name, need and contact details, asked one at a time and saved to a record you own. Nobody has to remember to follow up from memory.
Your documents on one side, a customer's question on the other, and a slot held on your real calendar before they close the tab.
Read, chunked and indexed automatically — no folders, no tags.
Private to you. Never used to train shared models.
Pricing sheets, refund policies, SOPs, contracts, onboarding docs — and the answers you've already typed into email a hundred times. Drop them in, forward them to your private capture address, or pick them out of Drive. Uploads land instantly and process in the background.
Every page is extracted, split into passages and indexed — not just filed under a title. Dates, amounts and terms are pulled out, so the assistant can quote your actual policy instead of paraphrasing the gist of it.
Two searches run on every question — one for meaning, one for exact words — and the results are fused, so “do you take Delta?” and “insurance accepted” both land on the same page of your policy. Every answer shows the file it came from. When nothing relevant comes back, it says so instead of improvising.
When someone's ready, the assistant checks your real Google Calendar free/busy, holds the slot, and emails a confirmation to both of you. Not a link to your calendar — the booking itself, on the calendar you already live in.
Every question it couldn't answer well is logged and grouped by topic. So you don't guess at what to add next — you get the list of what your customers keep asking that your documents don't cover, and answering one turns it into knowledge on the spot.
Every search is filtered by your account id inside the database query itself — not by an instruction in a prompt that another prompt could argue with. Never used to train shared models.
This started as a memory for your own documents, and it still is one. Insurance renewals, licences, registrations, filings — the assistant answering your customers is the same one that emails you before something quietly expires.
RegenticFlow is new. We have no customer logos to put here and we're not going to invent any. Here is what we can actually prove instead.
Every search is filtered by your account id inside the database query itself — not by an instruction in a prompt that another prompt could talk its way around.
Your documents and your customers' conversations are used to produce your answers. That is the entire list of what they're used for.
Export everything as JSON in one click. Delete your account and it cascades — documents, conversations, assistants, leads.
The widget only runs on domains you allow. Public conversations are gated by a challenge and a short-lived token bound to that assistant and that visitor.
The full version lives on our security page. Ask us where the edges are and we'll tell you — get in touch.
Publish it at your own link, drop one line of script into your website, or let people talk to it out loud. Same knowledge, same answers, same calendar behind all three.
It answers only from the documents you gave it, shows the source under each reply, and says it doesn't know rather than improvising when nothing relevant comes back. It's still a language model, so it can be wrong — which is exactly why every answer carries its citation and why you get a list of every question it fumbled, grouped by topic.
Upload what you already have — pricing sheet, refund policy, service list, the answers you've typed into email a hundred times — and publish. It's exactly as good as what you feed it: a five-page policy PDF makes a five-page-good assistant. The gap report tells you what to add next, and answering a gap turns your reply into knowledge on the spot.
No shared-model training, ever. Isolation is enforced in the database query — every search is filtered by your account id at the index, not by an instruction in a prompt. Export everything as JSON or delete the account and it cascades. The honest part: your content is processed by a third-party model provider under a no-training arrangement. We don't run our own model, and we're not going to pretend otherwise.
Nobody you've heard of. We're new, we have no logos to show you, and we're not going to invent any. What we'll do instead is let you build the whole thing free, publish it, and watch a real customer talk to it before you pay anything.