HackMIT 2025

Join us at MIT for an electrifying weekend at one of the world’s biggest hackathons.

September 13, 2025 to September 14, 2025

Kresge Oval

Introduction

Fetch.ai’s vision is to create a open AI Agent marketplace. We are empowering developers to build on our platform that can connect services and APIs without any domain knowledge.

Our infrastructure enables ‘search and discovery’ and ‘dynamic connectivity’. It offers an open, modular, UI agnostic, self-assembling of services.

Our technology is built on four key components:

uAgents - uAgents are autonomous AI agents built to connect seamlessly with networks and other agents. They can represent and interact with data, APIs, services, machine learning models, and individuals, enabling intelligent and dynamic decision-making in decentralized environments.

Agentverse - serves as a development and hosting platform for these agents.

Fetch Network - underpins the entire system, ensuring smooth operation and integration.

ASI:One - A Web3-native large language model (LLM) optimized for agent-based workflows.

Challenge statement

Bridge Intent and Action: Build AI Agents That Think, Plan, and Deliver

We live in an age of infinite possibilities but finite execution. People dream of financial independence, seamless automation, better health, smarter learning, or just getting more done. Yet most tools today—digital assistants, automators, and even LLMs—fall short of turning complex, human intentions into coordinated action.

This hackathon challenges you to build autonomous agents that understand what users actually want to achieve—and orchestrate the full stack of actions needed to make it happen.

Using the uAgents framework or any other agentic framework of your choice, create agents that can understand open-ended, natural language goals and break them into multi-step execution plans. These agents should verify outcomes and adapt in real time to ensure successful intent fulfillment.

Once built, register your agents on Agentverse, Fetch.ai's open marketplace where these agents can be discovered by other agents and users. Implement the Chat Protocol to to enable seamless natural language interaction via ASI:One, allowing users to talk directly to your agents.

From optimizing a user's portfolio to automating workflows or building playful tools that anticipate needs, your agents should go beyond commands to deliver outcomes.

This is your chance to shape the future of AI from passive tools to autonomous, intelligent, goal-driven agents.

Check out the resources to learn how to build and deploy your own AI agents.

What to Submit

Code
- Share the link to your public GitHub repository to allow judges to access and test your project.
- Ensure your README.md file includes key details about your agents, such as their name and address, for easy reference.
- Mention any extra resources required to run your project and provide links to those resources.
- All agents must be categorized under Innovation Lab.
  - To achieve this, include the following badge in your agent’s README.md file:
```
![tag:innovationlab](https://img.shields.io/badge/innovationlab-3D8BD3)
```
Video
- Include a demo video (3–5 minutes) demonstrating the agents you have built.

Tool Stack

Quick start example

This file can be run on any platform supporting Python, with the necessary install permissions. This example shows two agents communicating with each other using the uAgent python library.
Read the guide for this code here ↗

from uagents import Agent, Bureau, Context, Model
    class Message(Model):
        message: str
    
    sigmar = Agent(name="sigmar", seed="sigmar recovery phrase")
    slaanesh = Agent(name="slaanesh", seed="slaanesh recovery phrase")
        
    @sigmar.on_interval(period=3.0)
    async def send_message(ctx: Context):
       await ctx.send(slaanesh.address, Message(message="hello there slaanesh"))
        
    @sigmar.on_message(model=Message)
    async def sigmar_message_handler(ctx: Context, sender: str, msg: Message):
        ctx.logger.info(f"Received message from {sender}: {msg.message}")
    
    @slaanesh.on_message(model=Message)
    async def slaanesh_message_handler(ctx: Context, sender: str, msg: Message):
        ctx.logger.info(f"Received message from {sender}: {msg.message}")
        await ctx.send(sigmar.address, Message(message="hello there sigmar"))
        
    bureau = Bureau()
    bureau.add(sigmar)
    bureau.add(slaanesh)
    if __name__ == "__main__":
        bureau.run()

Important links

How to create an Agent with uAgents Framework ↗Communication between two uAgents ↗Communication between two uAgents using Chat Protocol ↗ASI:One API ↗How to create ASI:One compatible uAgents ↗How to write a good Readme for your Agents ↗

Video introduction

Introduction to agents

On Interval

On Event

Agent Messages

Examples to get you started:

Guide

How to Connect a Local Agent to Agentverse ↗

Guide

How to pass files to your uAgent ↗

Guide

MCP Integration Example ↗

Guide

Read our Medium Articles ↗

Judging Criteria

Functionality & Technical Implementation (25%)
- Does the agent system work as intended?
- Are the agents properly communicating and reasoning in real time?
Use of Fetch.ai Technology (20%)
- Are agents registered on Agentverse?
- Is the Chat Protocol implemented for ASI:One discoverability?
Innovation & Creativity (20%)
- How original or creative is the solution?
- Is it solving a problem in a new or unconventional way?
Real-World Impact & Usefulness (20%)
- Does the solution solve a meaningful problem?
- How useful would this be to an end user?
User Experience & Presentation (15%)
- Is the solution presented clearly with a well-structured demo?
- Is there a smooth and intuitive user experience?