Opengeni cloud
Sign in and go. You pay model cost plus 5%.




One dropped connection and the run is gone.
Agent code can't run next to your secrets.
Every user needs their own OAuth tokens.
Every API needs wiring before an agent can use it.
Agents forget everything between sessions.
Every query has to know who's asking.
A better model ships. You're locked in.
stripe.invoices.list3 invoices founddetect_duplicateduplicate confirmedstripe.refunds.createrefund issuedOur app is one client of the same API.
The same session, in your brand.
import { OpenGeniClient } from "@opengeni/sdk";
import { OpenGeniProvider, SessionConversation } from "@opengeni/react";
import "@opengeni/react/compiled.css";
// Talks to your backend. The API key never reaches the browser.
const client = new OpenGeniClient({ baseUrl: "/opengeni-api" });
export function BillingAssistant({ workspaceId, sessionId }: Props) {
return (
<OpenGeniProvider client={client} workspaceId={workspaceId}>
<div className="acme-agent" data-og-theme="light">
<SessionConversation sessionId={sessionId} />
</div>
</OpenGeniProvider>
);
}.acme-agent {
--og-color-accent: #5b4bff;
--og-radius-md: 6px;
--og-font-sans: "Archivo", sans-serif;
}A few lines. Your backend keeps the key.
.my-agent {
--og-color-accent: #5b4bff;
--og-radius-md: 10px;
--og-font-sans: "DM Sans", sans-serif;
}
<div class="my-agent" data-og-theme="light">Sign in and go. You pay model cost plus 5%.
Our Helm chart on any Kubernetes, with Terraform for AWS, Azure and GCP.
$ git clone https://github.com/Cloudgeni-ai/opengeni.git