Gatebreaking

Inspiration

Fashion inspiration is everywhere, but turning a saved outfit into something you can actually buy is slow and fragmented. Gatebreaking removes that bottleneck.

What it does

Upload an outfit image, and Gatebreaking identifies each clothing item, searches live shopping listings, and ranks the closest purchasable matches.

How we built it

We built a multi-agent pipeline using Fetch.ai uAgents and Agentverse. Claude analyzes outfit images, SerpAPI searches Google Shopping, and specialized search and ranking agents return the best matches through an orchestrator.

Challenges we ran into

Our biggest challenges were handling inconsistent model-generated JSON, search latency, coordinating asynchronous agents, and separating ASI:One user addresses from inter-agent Chat Protocol addresses.

Accomplishments that we're proud of

We created an end-to-end system that turns visual inspiration into ranked, shoppable product links through a coordinated multi-agent workflow.

What we learned

Reliable agent systems need more than strong models. Clear message contracts, request tracking, address handling, observability, and graceful failure handling are equally important.

What's next for Gatebreaking

Next, we want to enable automatic payments through Agentverse, allowing agents to complete purchases directly with merchant stores.

Built With

• claude
• fetchai
• python