[GPT-5 Compatible] AI Judgment: A Prompt that Analyzes FX Auto-Trading of Signal Tools at a Glance

A Prompt that AI analyzes and scores performance based on automated trading trade history is being released. It also supports the latest ChatGPT GPT-5.

Using this prompt, you can automatically generatevisualizations + AI evaluation reports like the following:

✅ How many wins and losses?
✅ Is the win rate and risk-reward favorable?
✅ What is the maximum drawdown?
✅ Are there trends such as "big win or steady win"?
✅ AI judgments and improvement suggestions as well!

What is a Prompt?

A prompt is an instruction sentence that directs the output direction and content for generative AI. By preparing a well-structured prompt in advance, you can efficiently obtain the targeted analysis results and graphs.

? Target audience

• Those developing automated trading tools
• Those looking to streamline signal verification
• Those who want to objectively understand FX trading performance
• Those who want to evaluate MT4/MT5 backtests

? Data required for analysis (CSV)

Just provide two CSV files used by AI judgment.

✨ Contents of the Auto-Analysis Prompt

The performance of automated trading will beanalyzed visually and quantitativelyas follows.

【Step 1: Statistical Summary】

Each item is quantified.

• Trading period (start to end)
• Total trades, total profit/loss (in yen)
• Winning trades profit/loss, losing trades loss
• Win rate (%)
• Profit factor (PF = total profit / total loss)
• Average risk-reward (average win / average loss)
• Average profit/loss (total P/L ÷ number of trades)
• Average pips (price difference × 100) with 1 lot = 100,000 currency units
• Maximum drawdown (yen and %)

【Step 2: 5 Graphs Display】

The following graphs are output.

• Cumulative Profit: trajectory of cumulative profit
• Trade-wise Profit: profit bars for each trade
• Profit Distribution: profit/loss distribution (with KDE)
• Win/Loss Category Breakdown: four-category pie chart of big win, small win, small loss, big loss
• Max Drawdown per Trade: drawdown trend per trade

【Step 3: AI Judgment (3 evaluation dimensions)】

Evaluate A to C on the following three aspects and provide brief comments for each.

• Stability of win rate (e.g., low variance, stable)
• Volatility handling (e.g., drawdown contained during rapid changes)
• Risk management (e.g., loss control, appropriate lot sizing)

【Step 4: Issues and Improvement Suggestions】

Based on the analysis results, present three concrete improvement points with issues and具体 improvements.

---

✅ Execution Result Image (Partial)

＜Sample Output＞

? Full Prompt

Please paste the following prompt into a generation AI such as ChatGPT. If you run it with attached CSV files, AI-generated analysis reports and graphs will be displayed on the screen.

You are an FX automated trading analysis assistant.
Read two CSV datasets (trade history and time-series chart) and output a self-contained HTML report in one file.
Graphs are created as PNGs and embedded directly in HTML via base64.

## Input Data (actual column names)
history.csv (trade history)
  open time
  open price
  close time
  close price
  order type (BUY/SELL)
  order volume
  profit and loss
  swap (optional)
  commission (optional)

bars.csv (time-series chart: OHLCV)
  time
  open
  high
  low
  close
  volume

## Step 1: Statistical Summary (Japanese)
Please present the following in Japanese as "item: value".
Trading period (start to end)
Total trades, total profit/loss (yen)
Winning trades profit/loss, losing trades loss
Win rate (%)
Profit factor (total profit / total loss)
Average risk-reward (average win / average loss)
Average profit/loss (total profit/loss ÷ number of trades)
Average pips (assume 1 pip = 0.01, calculated as BUY=+, SELL=-)
Maximum drawdown (yen and %)

## Step 2: Graphs + Comments (5 types)
1. Cumulative Profit (cumulative P/L line)
2. Trade-wise Profit (profit/loss bars per trade)
3. Profit Distribution (profit/loss histogram + KDE)
4. Win/Loss Breakdown (big win, small win, small loss, big loss pie chart)
   Note: Big win/big loss threshold is top 25% by |profit|
5. Max Drawdown per Trade (DD per trade)
All text inside graphs should be alphanumeric
Provide a brief Japanese comment under each graph

## Step 3: AI Judgment (All in Japanese)
Evaluate the following three aspects with A–C, and provide brief comments.
Win-rate stability
Volatility responsiveness
Risk management
Finally provide overall grade (A–C) and summary comment.

## Step 4: Issues and Improvement Suggestions (All in Japanese)
Based on analysis results, list three points in the following format.
Issue (concise heading)
Content (problem based on analysis)
Improvement (specific methods or approaches)

## Output format: HTML (one file)
HTML structure: ……
CSS supports Japanese fonts, full width, mobile-friendly
Graphs are embedded as base64 PNGs; no need to save
All text in Japanese
Do not output any extra notes or source references at the end

---

AI Judgment is a powerful tool to objectively review your automated trading strategy. We will continue to introduce analysis templates and use cases, so please keep using it.

▼ Products compatible with AI Judgment

【Trading Practice】 One-Click FX Training MAX

One-Click FX Arrows Auto Arrow Trading【AutoArwTrade】 support

One-Click FX Verification with AI【Test it Now】

Practice freely with a Completely No-Risk Trade Simulator!

Details page for One-Click FX Training MAX