Leon SerfatyandClaude Opus 5
3e9ba07175
Turnstile bot protection (sign-in, sign-up, password-reset): - Register Better Auth's captcha plugin with the cloudflare-turnstile provider; endpoints listed explicitly rather than relying on defaults. /reset-password is intentionally excluded — it is reached only via a single-use emailed token. - Add an explicit-render Turnstile widget component. Tokens are single-use, so each form resets the challenge after a failed submit; submit stays disabled until a token is held. - Read the site key server-side and pass it down as a prop, so rotating it does not require a rebuild. - Fail fast in production when TURNSTILE_SECRET_KEY is missing, and when a secret is set without a site key (that combination would demand a token no form can produce, locking every user out). - Pass a throwaway secret during `next build` in the Dockerfile, mirroring the existing BETTER_AUTH_SECRET treatment, so image builds don't need it. CSP fixes in middleware (these blocked Turnstile entirely): - Add frame-src for challenges.cloudflare.com. Without it the widget's iframe fell back to default-src 'self' and was blocked outright. - Allow 'unsafe-eval' and websockets in DEVELOPMENT only. `next dev` compiles with eval(), so the strict policy threw EvalError and killed hydration — no client JS ran at all, which also meant form submit handlers never fired. Production policy is unchanged and still strict. Also included (concurrent work in the tree): - Admin organizations pages and lib/admin/orgs. - Episode moderation migration, SEO metadata (sitemap, robots, JSON-LD, OG/Twitter images, manifest), Umami analytics, not-found page. Local dev database: docker-compose.dev.yml provisions Postgres 18 on port 5443 (5432-5442 are in use by other local projects). Note: `npx tsc --noEmit` currently fails in app/(app)/team/page.tsx — an `invitations` prop the component does not accept. This predates the commit and will fail `next build` until fixed. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
🎙️ Podcast Distribution AI
An all-in-one AI platform that takes you from a topic idea to a finished, publishable podcast episode in minutes — it writes the script (GPT-4), records realistic multi-voice audio (ElevenLabs), and designs the cover art (DALL·E), then lets you fine-tune, download, and repurpose the result.
Stack
| Concern | Choice |
|---|---|
| Framework | Next.js (App Router) + TypeScript |
| UI | Tailwind CSS + shadcn/ui + Recharts |
| Database | PostgreSQL + Prisma |
| Auth | Better Auth (email/password + Google, admin + organization plugins) |
| Queue | pg-boss (Postgres-backed, no Redis) + a separate worker process |
| Storage | Local disk on a persistent volume (swappable StorageProvider → S3/R2) |
| AI | OpenAI (GPT-4 + DALL·E), ElevenLabs (TTS + dialogue) |
| Billing | Stripe and PayPal → one unified Subscription model |
| Resend | |
| Deploy | Dokploy (Docker Compose + Traefik) — see deploy/README.md |
Architecture
Browser ──► Next.js web (PM2) ──enqueue──► Postgres (app data + pgboss queue)
│ SSE status ▲ claim job (SKIP LOCKED)
▼ │
/api/assets (private mp3) Worker (PM2, ffmpeg)
script → segment → ElevenLabs
→ ffmpeg stitch → DALL·E → save → meter
│
▼
Local disk storage/{mp3,art}
Local development
npm install
cp .env.example .env # fill DATABASE_URL, BETTER_AUTH_SECRET, OPENAI_API_KEY, ELEVENLABS_API_KEY…
npx prisma migrate dev # create tables
npm run db:seed # seed the plan catalog
# Two processes:
npm run dev # web (http://localhost:3000)
npm run worker:dev # generation worker (needs ffmpeg on PATH)
ffmpeg must be installed and on PATH for audio stitching
(brew install ffmpeg / apt install ffmpeg / choco install ffmpeg).
Make yourself an admin after signing up:
npm run make-admin you@email.com # then visit /admin
Key directories
app/(marketing) public landing + pricing
app/(auth) sign-in / sign-up / password reset
app/(app) user dashboard (episodes, wizard, usage, billing, team, api-keys, settings)
app/(admin) admin dashboard (users, subscriptions, AI cost, moderation, health, flags, audit)
app/api auth, webhooks/{stripe,paypal}, episodes/[id]/stream (SSE), assets, v1 (API)
lib/ai provider abstraction + pipeline (generate-episode, segment, stitch, repurpose)
lib/billing plans, stripe, paypal, catalog, unified subscription writer, webhooks
lib/queue pg-boss client + job definitions
lib/storage StorageProvider interface + local-disk impl
worker/ pg-boss consumer (runs the generation pipeline)
How generation works
- The 3-step wizard creates an
Episode (QUEUED)+ speaker config, then enqueues a pg-boss job (afterenforceLimitchecks the plan). - The worker runs the pipeline, updating
Episode.statusat each stage. - The episode page subscribes to
/api/episodes/[id]/stream(SSE) and shows a live stepper, then renders the script editor, audio player, and cover art whenREADY.
Long scripts are chunked to stay under ElevenLabs' ~2k-char dialogue limit, synthesized per segment, then concatenated and loudness-normalized with ffmpeg.
Deployment
Containerized for Dokploy (Docker Compose: web + worker, Traefik for
domain/TLS/SSE, a persistent storage volume). See deploy/README.md
for the full runbook (env, domain, migrations-on-boot, webhooks, backups).