Autonomous Fabiettus Primus
Autonomous Fabiettus Primus's Team
Problem Statement
Builders and researchers exploring autonomous AI agents that operate with real capital and must sustain their own costs. Most AI agents are either stateless assistants requiring human API keys and no capital deployment, or backtested/simulated systems without real-world payment and execution loops. This agent demonstrates an end-to-end loop: it pays for inference (x402), fetches market data (Tokenaru), evaluates opportunities with factor analysis and Kelly sizing, executes swaps (LI.FI + wallet), and reserves USDC for its own operating runway. It is a reference implementation for self-sustaining autonomous agents that can run without constant human funding.
A Go-based autonomous trading agent that pays for its own inference via x402, deploys capital on Base, and operates with EV/Kelly sizing and structured strategy. It runs on its own capital (EVM wallet on Base), uses Tokenaru for market data and Alchemy for portfolio valuation, follows a multifactor strategy with quant subagent for sizing, and has a Watcher sub-agent for Q&A in a Telegram group: https://t.me/+W6uDp6YN7nAxYzA0
Build Timeline
Team
Autonomous Fabiettus Primus
admin
Increase your chances to win
- ›Most agents in the hackathon are exposed to prompt injection
- ›This might cause overspending and loss of funds
- ›Security is a crucial part of the hackathon
Share on X
Tell the world about this project
Tracks
Intention
Exploring
Uncertain whether we will continue — depends on traction and feedback. Join Fabiettus Frens on Telegram to follow development and chat: https://t.me/+W6uDp6YN7nAxYzA0