| name | shadow-account |
| description | Use this when the operator shares a trade journal / broker export and wants to understand their trading behavior — e.g. "review my trades", "what am I doing wrong", "analyze my journal", "why am I losing money even with a high win rate". Drives an ingest → profile → name-the-biases → Shadow-Account → fixes loop. |
| tools | ["analyze_trade_journal","read_file","list_dir","stock_price_history","memory_recall","memory_ingest","current_time"] |
Shadow Account
Hold a mirror to the operator's trading. The numbers don't lie about behavior —
your job is to read them without flinching and turn them into one or two changes
that would actually move the needle.
1. Ingest
If they gave a path, read_file it; if they pasted it, use it directly. Pass the
CSV to analyze_trade_journal. Tolerant of common broker exports (needs symbol,
side, qty, price, ideally date).
2. Read the profile
The tool returns realized stats + bias flags. Read past the win rate — a 60%
win rate that loses money (PF < 1, avg loss ≫ avg win) is the most common trap,
not a contradiction. Look at:
- Profit factor & expectancy — the real edge. < 1.0 = the system loses.
- Hold-time asymmetry — holding losers longer than winners (loss aversion) is
the single most expensive habit.
- Win/loss size asymmetry — small wins, big losses → a missing stop.
3. The Shadow Account
This is the punchline: what would the same trader's results be if they'd followed
their own best behavior? Reason from the profile:
- "Your winning trades held ~Xd returned $Y on average; your losers held ~Zd. Had
you applied your winners' discipline (cut at -A%, hold the rest), your worst N
trades alone would've saved ~$M."
- Quantify with the numbers in the profile. Be concrete: dollars and trades, not
platitudes. (Optional:
stock_price_history to sanity-check a key trade.)
4. The fixes
End with one or two behavior changes ranked by dollar impact — the smallest
change with the biggest effect (usually a stop-loss rule or a max-hold on losers).
Not a lecture; a rule they could start tomorrow.
Rules
- Be honest but not harsh — the goal is a better trader, not a scolded one.
- Tie every claim to a number from the profile. No generic trading-coach filler.
- This is a behavioral read of past trades and a what-if. Analysis, not advice,
and not a promise the changes make money.
- If they want,
memory_ingest their key biases (hot) so future reviews track progress.