A local-first stock research toolkit for market data adapters, indicator engines, signal parsing, validation gates, and reinforcement review.
stock-kit turns local market data and public signals into testable research primitives: fetch, store, screen, plan, validate, and review.
Designed for software demonstration and research tooling. It does not place trades and is not financial advice.
AkShare-backed market data adapters feed a local SQLite repository with freshness checks.
Technical, volume, VCP, CANSLIM, and Wyckoff helpers create reusable signals for ranking.
Optional X and Reddit capture modules parse public text without account-changing automation.
Auction, intraday, post-close, scorer, and tuner modules keep decisions traceable.
python -m pip install -e ".[dev]"
python -m pytest
stock-kit --help
stock-kit fetch daily --tickers 000001,600519