YichengYang-Ethan

YichengYang-Ethan / oracle3

Public

AI-native prediction market trading agent on Solana, Polymarket, and Kalshi — HackIllinois 2026

github.comYichengYang-Ethanoracle3#readme ai-agent algorithmic-trading dflow kalshi llm
17
0
100% credibility
Found Mar 05, 2026 at 17 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Oracle3 is an autonomous AI-powered trading agent for prediction markets on Solana/DFlow, Polymarket, and Kalshi, with live dashboards, backtesting, risk management, and on-chain reputation tracking.

How It Works

1
🔍 Discover Oracle3

You hear about a smart trading helper that watches markets, spots opportunities, and trades prediction events automatically on Solana and other platforms.

2
📦 Set it up easily

Download and install with simple steps, no complicated tools needed, and browse live prediction markets right away.

3
🧪 Test with paper money

Run pretend trades using real market data to see how it performs, with a colorful dashboard showing profits and decisions in real time.

4
📊 Watch it learn and adapt

Tune strategies, backtest on past data, and watch your virtual portfolio grow as the agent spots edges across exchanges.

5
Go live safely
💰
Paper trading

Keep practicing risk-free while monitoring everything.

🔒
Live trading

Start with real money, protected by on-chain checks and MEV shields.

✅ Autonomous success

Your agent runs hands-free, building reputation on the blockchain while you relax and watch profits roll in.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 17 to 17 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is oracle3?

oracle3 is a Python-based AI-native trading agent for prediction markets on Solana (via DFlow), Polymarket, and Kalshi. It runs fully autonomous on-chain: fetching live data, generating signals via LLMs or quant strategies, managing risk with Solana simulations, and executing atomic trades—no human intervention needed. Users get a CLI for paper/live trading, real-time dashboard with equity curves, backtesting on historical episodes, and shareable Solana Blinks from the HackIllinois 2026 project.

Why is it gaining traction?

It packs eight production-ready on-chain features like cross-market arbitrage, Jito MEV protection, flash loans, and agent reputation scoring into a single agent framework, all verifiable via Solana memos. The live dashboard and Textual TUI make monitoring intuitive, while multi-provider LLM support (OpenAI, DeepSeek) lets devs swap models easily. Cross-platform price detection across exchanges hooks prediction market traders chasing arb edges.

Who should use this?

Solana DeFi builders prototyping autonomous agents for prediction markets. Quant traders backtesting momentum or orderbook strategies on Kalshi/Polymarket data. AI-native github devs experimenting with LLM-driven trading bots that sign real transactions.

Verdict

Grab it for agent experimentation—demo.sh spins up a full session in seconds, with strong docs and pytest coverage. At 17 stars and 1.0% credibility, it's early-stage (hackathon roots), so paper trade first before live Solana deploys. Solid reference for ai-native prediction market agents in 2026.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.