// the find
ivopetiz/algotrading
Algorithmic trading framework for cryptocurrencies.
A Python framework for building crypto trading bots against exchange APIs, with realtime, tick-by-tick, and backtest modes. Aimed at hobbyist quant traders who want to prototype simple SMA-crossover style strategies before risking real money, not people needing multi-exchange or production-grade execution.
Entry/exit strategies are plain functions that return True/False (see entry.py/exit.py) — no DSL or config schema to learn, and you can test a new signal in a few lines. The three operating modes share the same strategy code, so a strategy validated in backtest runs unchanged in realtime. There's an actual test/ directory covering aux, entry, exit, and finance modules, which is more than most solo-dev trading repos bother with. Stop loss and trailing stop loss are built in rather than left as an exercise for the user.
Despite the generic 'crypto exchanges' framing, the only exchange implementation in the tree is lib_bittrex.py — Bittrex shut down/restricted access years ago, so out of the box this doesn't connect to a live market without someone writing a new adapter. Historical market data (CSV and HDF files under hist-10m/hist-10s) is committed straight into the repo instead of gitignored, which is a sign the project hasn't separated code from data artifacts. The Travis CI badge points at travis-ci.com, which stopped free OSS builds years ago, so that pipeline is almost certainly not actually running despite the README implying active CI. The README itself flags that buy/sell order code is commented out and 'should only be used if you know what you are doing' — live execution isn't a supported, tested path, just a fend-for-yourself footnote.