Trade Journal (trading-journal)
The memory of the trading system. SQLite at ~/.trading-skills/journal.db.
trading-broker writes to it automatically; this skill is for everything manual.
python3 scripts/journal.py plan --symbol AAPL --side long --entry 195 --stop 188 --target 212 --qty 70 \
--setup "VCP" --thesis "3rd contraction, volume dry-up; wrong below 188"
python3 scripts/journal.py open --id 3 --price 195.4 --qty 70
python3 scripts/journal.py open --symbol MSFT --side long --price 410 --qty 20 --stop 398 --setup pullback
python3 scripts/journal.py close --id 3 --price 209.8 --reason target --fees 2.5
python3 scripts/journal.py note --id 3 --text "Held through the shakeout per plan" --grade A --tags patience
python3 scripts/journal.py update --id 3 --stop 196 # trailing the stop
python3 scripts/journal.py list --status open
python3 scripts/journal.py show --id 3
python3 scripts/journal.py import --file broker_export.csv # symbol,side,qty,entry_price,exit_price,entry_time,exit_time[,stop,fees,setup]
python3 scripts/journal.py export --file journal.csv
What to capture (prompt the user for it)
- Setup - a short consistent name (
VCP,pullback,ORB,earnings-gap) so analytics can compare setups. Reuse existing names (listshows them). - Thesis - why, and what would prove it wrong.
- Stop at entry - R-multiples are computed from the planned stop; without one, R is unknown.
- Grade (A-F) - how well the plan was followed, independent of P&L. A losing A-trade is fine; a winning F-trade is a warning.
- Notes after closing - one lesson per trade.
Notes
- R-multiple = net P&L / (|entry - planned stop| x qty). Fees are included in P&L.
- Moving the stop with
updatekeeps the original planned stop for R; this is deliberate. - Imported trades without a stop have no R; analytics still use their P&L.
- Use trading-review for statistics and coaching across many trades.