yuclaw 2.3.0

YUCLAW — Open Financial Intelligence Platform

YUCLAW — Open-Source AI Signal Research Platform

The open-source AI signal-research platform. Research and education only — not financial advice.

• Local LLM inference — Llama 3.1 70B (Q4_K_M) via Ollama on NVIDIA DGX Spark GB10, configured locally as the nemotron-3-super-local Ollama tag with a financial-analyst system prompt. Zero cloud LLM dependency. (Finnhub is the cloud data source for prices/news. Real Nemotron 3 Super 120B via OpenRouter is wired as a dormant fallback path.)
• Signal aggregator v2.4 — 6-component composite scoring over a 39-ticker universe (leveraged ETFs blocklisted).
• Strategy backtest engine — historical Calmar metric in output/backtest_all.json with documented limitations.
• Hash-anchored audit trail — selected signal-decision hashes anchored on Ethereum Sepolia testnet. See yuclaw-trust for the honest framing of hash-only vs. zk-SNARK.
• Macro regime detector — CRISIS / RISK_OFF / RISK_ON.
• Alpaca paper-trading bridge — yuclaw paper with seven safety nets: paper-URL guard, validation ping, market-hours check, $10K notional cap, first-run consent, append-only audit log, drawdown kill switch.
• Forward track record — daily entries built by cron/track_record_builder.sh after market close.

Install

pip install yuclaw
yuclaw today          # daily brief
yuclaw signals        # raw signal list
yuclaw regime         # macro regime
yuclaw brief          # LLM synthesis
yuclaw track          # forward track record (day N)
yuclaw dashboard      # open live dashboard
yuclaw paper          # Alpaca paper trading (with safety nets)

How it works

Market Data (Finnhub + yfinance, 39 tickers)
        |
        v
Factor Library  ->  Macro Regime  ->  Risk Engine
        |
        v
Signal Aggregator v2.4 (6-component composite)
        |
        v
Llama 3.1 70B (local, via Ollama)
        |
        v
Hash anchor on Ethereum Sepolia (selected signals)
        |
        v
Dashboard + Daily Brief + Forward Track Record

Methodology + limitations

The dashboard's BACKTEST RESULTS card is currently a placeholder pending live computation. The strategy backtest engine output (output/backtest_all.json) and the forward track record (output/track_record/dayN.json) both have documented limitations: no transaction costs, no slippage modeling, no bid/ask, no partial fills. See docs/methodology/backtest.md in the brain repo before drawing inferences from any single accuracy or Calmar number.

Disclaimer

YUCLAW is open-source research and educational software. It is NOT financial advice, investment advice, or a recommendation to buy, sell, or hold any security. All signals, scores, and analyses are generated by automated AI models and may contain errors.

Past performance does not guarantee future results. Trading involves substantial risk of loss. You are solely responsible for your own investment decisions. Consult a licensed financial advisor before making any investment.

YuClawLab, its contributors, and affiliates accept no liability for any losses arising from use of this software.

For educational and research purposes only. MIT Licensed.

Links

• GitHub: https://github.com/YuClawLab
• Dashboard: https://yuclawlab.github.io/yuclaw-brain
• Paper: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6461418
• Methodology: https://github.com/YuClawLab/yuclaw-brain/blob/main/docs/methodology/backtest.md
• Disclaimer: https://github.com/YuClawLab/yuclaw-brain/blob/main/DISCLAIMER.md