Fetching from the wire…
Public story · 2026-07-10 · high
Combined, the three projects have pulled in more than 160,000 GitHub stars, but none publish audited live trading returns.
Why now: The cluster surfaced in coverage dated July 10, 2026, with three trading-agent repos trending at once instead of the usual single standout.
Three AI trading agent projects landed on GitHub's trending list at once: HKUDS/Vibe-Trading, TauricResearch/TradingAgents, and ZhuLinsen/daily_stock_analysis.
TradingAgents has pulled in roughly 92,200 stars on GitHub, daily_stock_analysis about 56,500, and Vibe-Trading around 19,100. That's more than 160,000 stars spread across three repos, and star counts measure curiosity, not whether a strategy makes money.
Three showing up on the same day isn't a coincidence, it's a cluster. Finance keeps pulling in AI agent projects, and the reason probably isn't that language models are especially good at reading markets. It's that a backtest is cheap to generate and easy to make look convincing. Feed a model historical prices, let it write a strategy, run it against the past, and you get a chart that climbs up and to the right. Nobody has to risk a real dollar to publish that chart.
Vibe-Trading ships as a personal agent with backtesting built in, which is the feature that makes it shareable on GitHub, not the feature that makes it profitable. None of the three repos publish audited live trading results.
Backtested trading agents will keep trending on GitHub because backtests are the easiest kind of AI output to fake convincingly, not because the underlying strategies work. Watch for whether any of the three ever publishes forward-tested or live-money results instead of a historical curve fit.
Each link below shares sources, entities, or timing with this story.
The lab behind LightRAG (37,823 stars) and nanobot (45,882) landed correlation-regime detection plus a backtest/data/live-safety correctness pass on July 17, then Binance crypto fallback and parallel-execution fixes on July 18. Be skeptical of any LLM trading agent's claims. B...
HKUDS/Vibe-Trading (30,423 stars, MIT, v0.1.12 July 22) from Hong Kong University's data science lab pairs a reasoning loop of 50+ tools with 18+ historical data providers (Tushare, Yahoo Finance, OKX, CCXT) and nine market-specific backtesting engines covering US equities, A-...
Fundamental, sentiment, and technical analysts plus bull/bear researchers and risk managers over LangGraph, v0.2.5 adding GPT-5.5 coverage. Backtests claim 24.9% annual returns at a 5.60 Sharpe on AAPL/GOOGL/AMZN (GitHub). I'd take backtested Sharpe of 5.60 with heavy skeptici...
From HKUDS (the LightRAG lab), OpenHarness (Python, ~13,800 stars, v0.1.9) packages tool-use, skills, persistent memory, and multi-agent coordination, plus a personal assistant "ohmo" that talks to Slack and Feishu. It ships a 43-plus tool ecosystem with MCP support and unifie...
HKUDS/LightRAG (EMNLP 2025), microsoft/graphrag, and milvus-io/milvus remain the three anchors of graph and vector retrieval, all clustered within 10K stars of each other. None has broken away, which is itself the signal: graph RAG hasn't picked a winner after eighteen months....
Built by Bingxi Zhao in Chao Huang's HKUDS lab, it runs Chat, Quiz, Research, Visualize, Solve and Mastery Path on one unified agent loop: "you switch the objective, not the engine, and context moves with the learner." The memory design is the transferable idea. L1 is a worksp...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.