The SaaS boilerplate most AI agents prefer
Auth, payments, data and email in a type-safe architecture optimized for a clean context window and AI performance.
or see a demo →
Wired up, ready to go
Auth, payments, database, and email — configured and talking to each other from the first commit. Webhooks keep everything in sync automatically.
Every integration sits behind an abstraction layer, so you can swap providers without rewriting your app. Move from Clerk to Auth.js, Stripe to Lemon Squeezy, or Resend to SendGrid — your business logic doesn't change.
What you can build with it
We used A2SaaS to build AI-powered conversational analytics that monitor the health of your business. Not a chatbot — a data analyst.
Let me pull churn, retention, customer health, and runway…
Good news — no major threat right now. Churn is negligible (well under the 3% benchmark), NRR is solid at 103%, and runway is comfortable at ~17 months. One flag: customer growth slowed to 7.8% MoM — worth watching, though MRR growth is accelerating on expansion revenue.
Static dashboards show you numbers. This one talks back. Built on the same stack you get from npx create-a2saas.
Try the live demo →Built with the most trained-upon stack in open source
Every technology chosen for maximum AI compatibility and developer productivity. No exotic tools, no custom frameworks — just the proven ecosystem with the deepest representation in training data.
Schema in code, not config files. Ready-to-deploy components, blocks, and templates.
- src/
- app/
- (auth)/
- (marketing)/
- (app)/
- setup/page.tsx
- api/
- v1/
- webhooks/
- components/
- server/
- lib/
- env.ts
- proxy.ts
- app/
Built for the context window
Flat, scannable, zero magic
No hidden logic or deeply nested abstractions. The file structure is flat and the imports are obvious — so any AI assistant can read the whole project context and modify it accurately.
Latest Writing
View Archive →We Asked Our Dashboard What the Biggest Threat to Our Business Was
We built an AI analyst into our SaaS dashboard, then asked it one open-ended question. Here's what it pulled, what it said, and why conversational analytics beats staring at charts.
Static Dashboards Are a Solved Problem. What Comes Next?
Every SaaS product ships the same dashboard: MRR top-left, churn in the middle, a line going up and to the right. They're fine. They're also completely passive. Here's what we think replaces them.
CLAUDE.md: The File That Makes Your AI 10x More Effective
Context files like CLAUDE.md tell AI assistants how your codebase works before they write a single line. Here's why they matter and how to write a good one.
Start building
One command. Type-safe foundation. AI-native architecture.