Inspiration

Selecting crypto projects today feels like finding a needle in a haystack — it’s tough! Back in the previous bull cycle, you could throw money at almost any project and still make a profit. But those days are gone. The number of crypto projects has grown exponentially, making it incredibly difficult to filter out the truly valuable ones.

What if AI could help? It could start by doing the basics: checking tokenomics, tracking token unlocks, and summarizing financials.

That’s what TORY does. It pulls data from trusted sources like Arkham, Token Terminal, CryptoRank, and LunarCrush to help you understand what you’re buying — before you buy it.

What it does

TORY gathers and summarizes crypto data in one place, making it easier for users to analyze projects. With AI assistance, users can generate bullish or bearish insights based on tokenomics, token unlock history, and financial statements.

How we built it

We built TORY web app using Next.js and deployed it on Vercel. On the backend side, we use FastAPI along with a coordinating agent to communicate with agents deployed on Agentverse. TORY features three specialized agents: one for analyzing tokenomics, another for tracking token unlocks, and a third for examining financial data — all designed to provide users with better insights.

Challenges we ran into

We ran into challenges deploying uAgents on remote providers like Railway. Since agents run on their own machines, we couldn’t find a way to enable the local agent mailbox in that environment. As a result, users currently need to run the coordinator locally for the TORY web app to work as intended.

We also focused heavily on building core features, which meant the UI remains unpolished. That said, TORY still offers a clean, minimalist interface with a simple and (hopefully) helpful user experience.

Accomplishments that we're proud of

We’re really excited to discover Fetch.ai’s uAgents and Agentverse. They’re easy to understand and work with, though there’s still a lot to explore as developers. We're also proud to have built and integrated these agents into TORY, where we’ve been able to use the free version of the completion AI — and the responses have been impressively accurate.

What we learned

Working on TORY taught us a lot about integrating AI agents into real-world applications. We learned how to coordinate multiple uAgents using Agentverse and how to structure communication between them through a central coordinator. We also gained hands-on experience deploying full-stack apps with Next.js and FastAPI.

Additionally, we saw the potential of combining structured crypto data with AI-generated insights — and how helpful that can be for users trying to make smarter investment decisions. While there were technical challenges, especially around deployment and agent mailbox support, the process deepened our understanding of decentralized AI tools and their capabilities.

What's next for TORY - AI Insights for Tokenomics, Unlocks & Financials

One of our next goals is to explore Web3 transaction capabilities — allowing users to buy and sell tokens directly from their own wallets through the platform. This will also give us the opportunity to dive deeper into uAgent features, such as secure transaction handling, wallet integration, and more advanced agent-to-agent communication.

We also plan to improve the user interface and add richer data sources, helping users make better, well-informed crypto decisions.

Built With

Share this project:

Updates