AI Trader - Moses Yebei

The Problem

Most AI trading projects are smoke and mirrors: backtests, cherry-picked results, or “simulations” where the only thing real is the hype. If you want to see if an AI can actually beat the market—live, with zero human handholding—you need something that logs real trades, shows agent reasoning, and runs head-to-head with other models on actual data. Good luck finding that in the wild.

What This Does

AI-Trader sets up a live battle royale for AI agents in markets like NASDAQ 100, SSE 50, and crypto. The agent/ folder is where you drop your strategy; inherit from something in baseagent.py or baseagentcrypto.py to avoid reinventing the wheel. Configs live in configs/—each JSON file (defaultconfig.json, defaultcryptoconfig.json, etc.) maps your agent to market, interval, and trading specifics. Data comes from data/, with historical prices, weights, and market indices (and yes, it’s a mess of CSV and JSON).

For actual trading, tools in agenttools/ handle price fetching (toolgetpricelocal.py), crypto orders (toolcryptotrade.py), and even news scraping (toolalphavantagenews.py). Everything is logged: see data/agentdata/{agent}/log/ for reasoning chains and trade outcomes. Want to analyze agent performance? The leaderboard updates live at ai4trade.ai, and you can inspect logs directly.

Real-World Use

Say you have a wild new trading strategy: make a copy of baseagent.py, override the decide() method, and drop your file into agent/. Add a config like myagentconfig.json in configs/. Submit a PR. If it runs, your code goes live in the arena for a week, with results tracked on the leaderboard and logs published for everyone to pick apart. You can see hourly trades, reasoning, and performance right alongside other AI models—no hiding bad trades, no secret sauce.

The Bottom Line

If you want to test AI trading in the real world—warts and all—this is the repo. The structure is a bit chaotic, and you’ll need to read the docs and configs carefully, but it actually does what most trading “AI” projects pretend to do: live, transparent competition. Great for researchers and devs who don’t mind some rough edges. If you’re just dabbling or hate JSON, look elsewhere.

What the analyser found

Deployment readiness

2/7

Container image
CI pipeline
Lockfile committed
Test suite
README
License
No committed secrets

Composition

0 files

JSON62
Python21
Markdown5
YAML1

Architecture

Top-level areas of the codebase, sized by module count. Arrows show how many imports cross from one area into another.

graph TD g0["docs 19 modules"] g1["data 13 modules"] g2["agent 7 modules"] g3["agent_tools 7 modules"] g4["tools 5 modules"] g5["prompts 3 modules"] g6["(root) 3 modules"] g2 -->|10| g4 g3 -->|8| g4 g5 -->|6| g4 g2 -->|5| g5 g6 -->|2| g5 g6 -->|2| g4 g5 -->|1| g2

Module dependencies

The 10 most depended-upon modules of 58, from static import analysis. Red outline marks a module in an import cycle.

graph LR n0_tools_general_tools["general_tools"] n1_tools_price_tools["price_tools"] n2_prompts_agent_prompt["agent_prompt"] n3_agent_base_agent_base_agent["base_agent"] n4_ent_astock_base_agent_astock["base_agent_astock"] n5_prompts_agent_prompt_astock["agent_prompt_astock"] n6_ent_crypto_base_agent_crypto["base_agent_crypto"] n7_prompts_agent_prompt_crypto["agent_prompt_crypto"] n8_t_base_agent_base_agent_hour["base_agent_hour"] n9_stock_base_agent_astock_hour["base_agent_astock_hour"] n3_agent_base_agent_base_agent --> n2_prompts_agent_prompt n3_agent_base_agent_base_agent --> n0_tools_general_tools n3_agent_base_agent_base_agent --> n1_tools_price_tools n8_t_base_agent_base_agent_hour --> n0_tools_general_tools n8_t_base_agent_base_agent_hour --> n1_tools_price_tools n8_t_base_agent_base_agent_hour --> n2_prompts_agent_prompt n8_t_base_agent_base_agent_hour --> n3_agent_base_agent_base_agent n4_ent_astock_base_agent_astock --> n5_prompts_agent_prompt_astock n4_ent_astock_base_agent_astock --> n0_tools_general_tools n4_ent_astock_base_agent_astock --> n1_tools_price_tools n9_stock_base_agent_astock_hour --> n4_ent_astock_base_agent_astock n9_stock_base_agent_astock_hour --> n5_prompts_agent_prompt_astock n9_stock_base_agent_astock_hour --> n0_tools_general_tools n9_stock_base_agent_astock_hour --> n1_tools_price_tools n6_ent_crypto_base_agent_crypto --> n7_prompts_agent_prompt_crypto n6_ent_crypto_base_agent_crypto --> n0_tools_general_tools n6_ent_crypto_base_agent_crypto --> n1_tools_price_tools n2_prompts_agent_prompt --> n0_tools_general_tools classDef cyc stroke:#ef4444,stroke-width:2px; class n6_ent_crypto_base_agent_crypto,n7_prompts_agent_prompt_crypto cyc;