Backtesting.py

Backtesting.py is a small Python library. Give it a pandas table of OHLCV candles and a strategy class with entry and exit rules, and it returns the usual statistics (return, win rate, drawdown, Sharpe, exposure) and an interactive chart of every trade. It also has a parameter optimizer.

What it adds to Moby. A quick backtest on the exact data your Moby agent trades. It does not care where candles come from, so your agent can test a rule on a Solana token's own history, new launches included, once there is enough of it.

How to pair it. Turn a strategy idea (for example a breakout with Moby's default -30% stop and +100% take-profit) into a Backtesting.py strategy, run it on a few tokens from get_token_ohlcv, and show the stats before you save it as a Moby strategy. Moby returns candles newest-first; reverse them before you feed them in.

What it does not do. It tests one instrument at a time with a simple fill model, so slippage on thin Solana pools is understated; add a generous commission. It does not trade. Its license is AGPL-3.0, which matters only if you ship it inside a product.