SalesIntel AI

AI-native revenue intelligence for sales managers and RevOps. Next.js 16 · React 19 · TypeScript · Tailwind 4 · Supabase (Auth, Postgres + RLS, Storage) · Gemini.

Setup

1. npm install
2. Create a Supabase project and run supabase/migrations/0001_schema.sql in the SQL editor.
3. Copy .env.example to .env.local and fill it in (Supabase URL/keys, GEMINI_API_KEY, payment links).
4. In Supabase → Authentication → URL configuration, add http://localhost:3000/auth/callback (and your production URL) as redirect URLs.
5. npm run dev — visit /setup for a live configuration checklist.

Scripts

• npm run dev / npm run build / npm run lint / npm run typecheck
• npm test — metrics-engine tests and a database test that applies the migration to an in-process Postgres and checks workspace isolation, roles and credits.

How it works

• Isolation: every business table has org_id and RLS; roles (owner/admin/member) are enforced in SQL. Cross-workspace links are blocked by composite foreign keys.
• Intelligence: src/lib/intel computes metrics and the six findings deterministically from your records; Gemini (server-side only) narrates them. Rule-based Next Best Actions need no AI.
• Credits: spent atomically via a service-role-only SQL function and refunded if an AI call fails. Costs and plans live in src/config/pricing.ts (and the plans table).
• Payments: hosted payment links only (STARTER_PAYMENT_LINK, PRO_PAYMENT_LINK). Apply a plan after the provider confirms payment with POST /api/billing/activate (Bearer BILLING_ADMIN_SECRET).
• Limits: no email sending, no external lead database, audio ≤ 15 MB (inline to Gemini).