Spendwall

Inspiration

As AI agents become capable of making purchases and financial decisions, we realized there needs to be a safety layer between AI and our money. We wanted to build something that gives people the convenience of autonomous AI while keeping them in control of their spending.

What it does

SpendWall is an AI-powered financial firewall for agentic commerce. Users can create spending rules, set limits, and monitor transactions. SpendWall analyzes purchases in real time, identifies risky or unexpected transactions, and can flag or stop transactions that violate the user's rules. Its audit trail also explains what happened and why.

How we built it

We built SpendWall using React, Vite, Tailwind CSS, Node.js, Supabase, Groq AI, and a Chrome extension. The frontend provides the Control Center, transaction dashboard, checkout simulator, and AI Copilot, while the backend handles transaction processing and AI-powered analysis.

Challenges we ran into

One of our biggest challenges was connecting multiple technologies and making them work together reliably. We also had to figure out how to represent AI decisions in a way that was understandable to users while keeping the experience fast and intuitive.

Accomplishments that we're proud of

We're proud of creating a working prototype that demonstrates how AI agents and financial safety can work together. We especially like our real-time transaction monitoring, customizable spending rules, AI explanations, and audit trail.

What we learned

We learned how important it is to design for trust and transparency when building AI-powered systems. We also gained experience integrating AI, databases, frontend interfaces, and browser-based functionality into one product.

What's next for Spendwall

We want to expand SpendWall into a real financial security layer for agentic commerce. Future features could include smarter fraud detection, personalized risk models, bank integrations, mobile support, and more advanced controls for autonomous AI agents.