Public algo‑trading toolkit

A public GitHub repository is making the rounds as a map of algorithmic-trading software, grouping tools for backtests, optimization and live execution in one place. (github.com) Algorithmic trading means writing rules that decide when to buy or sell, then testing those rules on old market data before risking real money. The repository highlighted this week is PyTrade, a curated list on GitHub that says it covers Python packages and resources for “backtesting, live trading” and related workflows. (github.com) PyTrade organizes the stack by job: trading frameworks, backtesting frameworks, strategy examples, machine-learning tools, analytics, data providers, broker and exchange APIs, dashboards and research tools. Its README says the project is “inspired by PyViz” and presents itself as an “awesome-algo-trading list” with a more structured interface. (github.com) The list mixes traditional finance and crypto-native software in the same catalog. In its trading-framework section, PyTrade names QuantConnect’s Lean, NautilusTrader, Backtrader, PFund, Trading Strategy and Blankly, and describes PFund as supporting “TradFi+CeFi+DeFi.” (github.com) That blend reflects how trading infrastructure has shifted over the past few years. Open-source projects now pitch the same codebase for research and production, with “backtest-live parity” — using one engine in simulation and in live markets — as a selling point for reducing deployment errors. (nautilustrader.io) The optimization piece is central because most trading systems fail in the jump from a good historical test to a bad real-market result. Freqtrade, one of the projects commonly cited in these lists, says it includes backtesting, plotting, dry-run simulation and “hyperoptimization” to tune buy, sell, stop-loss and trailing-stop parameters before live trading. (freqtrade.io) The DeFi side solves a different problem: trading on blockchains and decentralized exchanges, where execution, fees and liquidity work differently from a stock broker. Trading Strategy’s Trade Executor describes itself as a Python framework for “backtesting and live execution” of algorithmic strategies on decentralized exchanges. (github.com) This is not the first attempt to catalog the field, but newer lists are getting denser and more explicit about maintenance. The “best-of-algorithmic-trading” project says it tracks 100 open-source projects with about 310,000 combined GitHub stars, updates weekly and ranks projects by a quality score built from GitHub and package-manager signals. (github.com) Older “awesome” lists leaned more heavily toward tutorials, papers and courses. The “awesome-algorithmic-trading” repository, for example, bundles strategy topics, academic papers, books and university coursework alongside software links, which makes it useful for learning but less focused on deployment plumbing. (github.com) The practical message in the latest roundup is narrower: the hard part is no longer finding a bot, but choosing a testing and execution stack that matches the market you want to trade. Public curation helps with discovery, but every major framework still warns users to test in simulation first and not treat historical results as a guarantee. (freqtrade.io)