chencore

chencore / autoresearch-crypto

Public
90
20
69% credibility
Found May 09, 2026 at 90 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

An open-source framework that uses AI agents to autonomously discover, evolve, backtest, and execute cryptocurrency trading strategies on DEX platforms like Nado.

How It Works

1
📖 Discover the crypto trading helper

You hear about a smart tool that finds and improves trading plans for cryptocurrencies like ETH automatically.

2
📥 Get it set up

Download the tool and prepare it on your computer in a few simple steps.

3
📊 Grab price history

Pull recent price charts for your favorite crypto so the tool can learn from real market moves.

4
🔍 Let AI hunt for winning plans

Watch as smart agents explore, compete, and evolve the best trading ideas tailored to current markets.

5
🧪 Test on past data

Run pretend trades using history to see how the plans would have performed safely.

6
Ready to trade for real?
🔄
Practice more

Tweak and retest plans until you're confident.

💰
Start live trading

Link to a crypto exchange and let the best plan trade automatically.

💹 Automated profits roll in

Your evolved trading plan works around the clock, turning market moves into steady gains.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 90 to 90 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 autoresearch-crypto?

autoresearch-crypto is a Python framework that automates quantitative trading strategy discovery for crypto perpetuals. It uses AI agents to evolve strategies via evolutionary algorithms, backtest them with realistic fees and slippage, and deploy live on Nado DEX or OKX. Developers get a full pipeline: download data, run quick strategy searches, backtest checkpoints, and fire up live bots with CLI commands like `uv run python live_nado_quant.py --ticker ETH --capital 100`.

Why is it gaining traction?

It skips manual hyperparameter tuning with agent-driven evolution engines that compete and reflect on performance, producing adaptive ensembles across 11 strategy types like trend, scalping, and hybrids. Live trading optimizes maker/taker orders for low fees, includes market regime detection and multi-timeframe signals, and handles real-world edge cases like pending orders. At 90 stars, it hooks devs tired of static backtesters by delivering production-ready crypto bots fast.

Who should use this?

Crypto quants prototyping strategies on ETH or SOL perps without coding each indicator from scratch. Python traders bridging backtests to live DEX execution on Nado or OKX. Devs researching autoresearch techniques who want walk-forward validation and risk-penalized scoring out of the box.

Verdict

Worth forking for crypto strategy research—installs cleanly with uv, runs searches in minutes, and ships live scripts. 90 stars and thin docs signal early stage; 0.7% credibility score means audit code and paper-trade first. Strong MIT-licensed base for custom bots.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.