ChainWatch AI

Inspiration The inspiration for ChainWatch AI came from the need for a decentralized system of intelligent agents that could assist with crypto-related tasks. With the increasing importance of real-time data, timely news, and automated alerts in the crypto world, we wanted to create a platform where multiple micro-agents collaborate to provide key insights and automated actions for crypto enthusiasts.

What it does ChainWatch AI is a suite of four specialized AI agents that work together to provide real-time crypto price data, aggregated news, dynamic alerts for price movements, and an assistant that integrates all agents. The agents communicate seamlessly to provide insights for crypto investors and enthusiasts, offering a personalized and efficient platform.

How we built it ChainWatch AI is built with FastAPI and uvicorn, utilizing microservice architecture to run each agent as an independent service. The agents interact via REST APIs, making it modular and easy to scale. The agents use mocked data for now, with plans to integrate live data in the future. Each agent was deployed individually, and they communicate through HTTP requests to handle various tasks, from providing price data to sending alerts.

Challenges we ran into One of the biggest challenges we faced was making the agents communicate with each other in a seamless manner while ensuring each was deployed correctly. We also faced challenges while integrating the Agentverse platform, especially handling the agent registration process. Another challenge was managing agent dependencies and ensuring they responded efficiently to requests.

Accomplishments that we're proud of We successfully built four distinct agents that each handle different tasks:

PriceAgent: Provides real-time mock crypto price data.

NewsAgent: Aggregates top crypto news headlines.

AlertAgent: Sends alerts when specific price thresholds are hit.

AssistantAgent: A dynamic assistant capable of interacting with all other agents on the platform.

These agents can run independently, but they also communicate with each other to form a cohesive system that can be used for both research and real-world crypto trading scenarios.

What we learned Through this project, we learned a great deal about how microservices and APIs work in practice. We also learned how to work with FastAPI and Agentverse to create scalable, modular systems that can be easily expanded. The most significant takeaway is understanding how to manage multiple services and ensure they integrate efficiently in a larger system.

What's next for ChainWatch AI Live Data Integration: We plan to replace the mocked data with real-time crypto prices and news feeds from APIs like CoinGecko or NewsAPI.

User Personalization: Adding more features to the AssistantAgent so it can track user preferences, suggest tailored alerts, and provide detailed insights.

Enhanced Notification System: Expanding AlertAgent to send notifications across multiple channels (email, SMS, etc.).

Scalability: Improving the system to handle higher volumes of users and adding more agents for other tasks like portfolio tracking and sentiment analysis.

Built With

• agentverse
• fastapi
• fetch.ai
• python
• uagents
• uvicorn