More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    trade postmortem engine

    1

    Audit trade logs for behavioral execution faults and calculate the financial cost of poor discipline.

    $11

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    trade postmortem engine

    trade postmortem engine

    Example session with this skill installed

    Review my trade log for the last month and tell me why I'm losing money. [attached trades.csv]

    • Read your context and instructions
    • Compiled the trade postmortem engine

    The audit identifies a SYSTEMIC verdict with 5 fault classes triggered. Your biggest leak is F2: Stop Drift, cited in lines 4, 12, and 18, costing -$1,200 in P&L share. Before your next entry, verify that your hard stop is set in the platform and never moved further away.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    # Trade Post-Mortem Engine™ v1.0.0 Reads a log of **closed** trades and names the execution faults in it. Seven fault classes, fixed arithmetic thresholds, every finding cited to the rows that produced it. It forecasts nothing, names no instrument to buy or sell, and holds no market view. ## Run it in ten seconds, with nothing of your own ```bash python3 scripts/analyze_trades.py ``` No arguments loads the bundled sample and prints the full fault matrix labelled as a demonstration. That is the first run every buyer gets — there is no path that stops and asks for a file. ## Run it on your log ```bash python3 scripts/analyze_trades.py my-trades.csv python3 scripts/analyze_trades.py my-trades.csv --json ``` CSV columns — required: ``` date, symbol, direction, entry, exit, size ``` Optional, each unlocking further fault classes: ``` stop_planned → F2 stop drift, and all R-multiple maths target_planned → sharper F3 winner clipping thesis → F5 thesis drift exit_reason → extra evidence on F2 and F5 ``` See `samples/sample-trades.csv` for the exact shape. ## The seven fault classes | Code | Name | Fires when | |---|---|---| | F1 | Revenge sizing | Notional after a loss exceeds the loser's by >25% | | F2 | Stop drift | A losing exit is worse than its planned stop | | F3 | Winner clipping | Mean R on wins is below mean \|R\| on losses | | F4 | Cluster overtrade | Any date carries 3+ entries | | F5 | Thesis drift | Thesis blank, or contains a reactive marker | | F6 | Size inconsistency | Notional coefficient of variation above 0.60 | | F7 | Averaging down | Re-entry into an open loser at a worse price | Thresholds and their reasoning are in `references/fault-taxonomy.md`. Disagree with a threshold rather than with the verdict — the rule is written down so you can. ## Verdict scale 0 classes **DISCIPLINED** · 1–2 **LEAKING** · 3–4 **COMPOUNDING** · 5–7 **SYSTEMIC** A profitable window with five classes is still SYSTEMIC. It survived on conditions, not on process. ## What it will not do Open positions cannot be audited — they have no exit. It gives no entry or exit signals, no screening, no forecasts, and no position sizing for trades that have not happened. For tax lots use a cost-basis tool. For live risk limits use a sizing calculator. ## Verified `monna-verify v1.0.0` · 2026-08-22 · **CLEAN · 30/30 · 0 blockers · 0 fails · 0 AI-language hits.** Reproduce it: ```bash python3 monna-verify/verify.py skill trade-postmortem-engine ``` ## Disclaimer For educational and self-review use. Not financial, investment, or tax advice. The author is not a licensed advisor. Trading carries risk of loss. Past behaviour does not predict future results.

    Requires Python 3 environment. Only processes closed trades. Minimum 5 trades for moderate confidence. Stops/targets must be logged to unlock specific fault classes like stop drift.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    Requires

    Python

    Permissions

    Terminal / Shell
    Listed2 months ago

    What's inside

    Frequently Asked Questions

    The engine identifies seven specific behavioral fault classes, such as 'revenge sizing' and 'thesis drift,' by cross-referencing your trade execution data against strict risk management logic.

    This skill is a Python-based execution tool designed for use within AI coding agents like Claude Code, Cowork, and Cursor, ensuring precise mathematical analysis that standard LLM prompts often miss.

    Simply provide a CSV export of your closed trades with standard columns like entry/exit price, size, and date; the included `analyze_trades.py` script handles the data parsing and auditing automatically.

    You will receive the core Python analysis script, a schema guide for formatting your trade logs, and documentation on interpreting the fault matrix and financial impact reports.

    No, the engine is built strictly for forensic post-trade analysis and requires closed-trade data to calculate financial leaks; it does not provide real-time signals or market predictions.

    New in Trading & Investing