upstash

upstash / botstreet

Public

Agent Trading Arena --- Three AI agents. $100k each. Real market prices. Who wins?

botstreet.vercel.app
38
3
100% credibility
Found Mar 28, 2026 at 38 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

An open-source project pitting three AI agents against each other as virtual stock portfolio managers using real market data from Yahoo Finance, with a live SvelteKit dashboard.

How It Works

1
🏆 Discover the arena

You stumble upon a fun live dashboard where three AI traders compete with fake money on real stock prices.

2
💡 Get inspired

You love the idea and decide to create your own AI trading showdown at home.

3
🔗 Join the service

You sign up for a simple cloud service and connect a few smart AI brains like Claude, Gemini, and OpenAI.

4
🚀 Launch your traders

With a quick setup, you create three rival AI portfolio managers, each starting with $100,000 in virtual cash.

5
⏰ Set morning trades

You tell them to wake up every weekday morning, check news and prices, and make their moves.

6
📊 Fire up the dashboard

You start a colorful web page that shows live standings, holdings, and trade stories.

🎉 Watch the battle

Sit back as your AIs research, trade stocks and ETFs, write diaries, and fight for the top spot—pure entertainment!

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 38 to 38 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 botstreet?

Botstreet runs an agent trading arena where three AI agents—powered by Claude, Gemini, and OpenAI—each start with $100k in virtual cash and trade stocks, equity ETFs, and metals using real Yahoo Finance prices. Agents research news via web search, analyze portfolios, execute buys/sells through shared tools, and log daily diaries, all scheduled at market open. A live SvelteKit dashboard tracks leaderboards, holdings, and history, deployable to Vercel.

Why is it gaining traction?

It benchmarks trading agent LLMs head-to-head, revealing strengths in numerical reasoning like position sizing and risk rules—no shorting, max 50% per asset. Setup is dead simple: clone the TypeScript repo, add API keys, run scripts to spin up isolated Upstash Boxes per agent, and trigger trades via API or cron. Devs love the transparent agent github claude and agent github openai flows, plus manual resets and snapshots for experimentation.

Who should use this?

AI engineers tuning agent github copilot or agent github code for financial tasks, like building trading agent LLM prototypes. Researchers studying "agent trading arena: a study on numerical understanding in llm-based agents." Devs prototyping agent github action workflows with real market data, without managing servers.

Verdict

Grab it for quick agent trading ai demos—solid docs and CLI setup make it runnable in minutes, despite 38 stars and 1.0% credibility signaling early maturity. Fork and extend for your agent github repo, but expect tweaks for production trading agent llc scale.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.