Log and review every trade decision in your Notion ledger

Trading Ledger is a Market Wizards-style trading journal: the user reports a fill in plain language ("bought 500 NVDA at 135, stop at 128, betting the post-earnings dip fills"), and the agent writes it into the user's own Notion database - ticker, size, price, plus what a spreadsheet journal loses: the Thesis, the Plan, and the Emotion at entry. A missing thesis is the one field worth interrupting for, since entry reasons decay overnight; anything else uncertain gets recorded as-is with the open question logged in Notes and the row marked To-confirm. It confirms the exact target Notion database by title and data_source_id before any write, using the schema the companion template ships: Market, Direction, Size, Entry/Exit Price, Entry/Exit Date, Thesis, Plan, Emotion (Calm/FOMO/Panic/Revenge/Boredom/Overconfidence - tagged only from what the user admits, never diagnosed), Execution (Per plan/Early exit/Delayed stop/Impulse/Unplanned add), P&L, and Status (Open/Closed/To-confirm/Reviewed). Closing a trade grades Execution against the user's own Plan, not against hindsight - a loss taken exactly at the planned stop is graded Per plan, a good trade, even though it lost money.

Use it when a user reports a trade fill, says "log a trade" or "tidy up my trading ledger," or asks to "review my trades." It never computes P&L the user hasn't given it (options and futures multipliers make guessing unsafe), never invents a select-field value outside the fixed enum, and never looks up market prices to fill a gap - the broker knows the fills, only the user knows the reasons, so self-report is mandatory by design; there is no broker integration.

Input is a plain-language trade report or a review request. Output is a Notion database row (or update) plus a short receipt, batching every open question into one message rather than asking one at a time. A review answers, per closed trade, whether the thesis played out, how the execution graded against the plan, and what share of trades were emotion-tagged, writing the conclusion into Review and moving Status from Closed to Reviewed; open positions get asked whether the entry thesis still holds today.

The official Notion connector (MCP) only - no shell commands, no network fetches, no market-data lookups, no credentials, and writes are scoped to the exact database ID the user confirmed for the session. Date fields must be expanded as "date:Entry Date:start": "YYYY-MM-DD" or the Notion API returns a 400; a known Notion MCP issue can silently drop that expanded date on create-pages, so the row is read back and patched with update-page if the date is empty.

Traders who want a real decision journal - thesis, plan, and emotion at the moment of entry - graded on decision quality rather than P&L, kept in their own Notion workspace instead of a spreadsheet that only records the fill.