- Home
- Skills
- Trading & Investing
- Trading Performance Forensics AI
Works with the AI Agents you already use
Trading Performance Forensics AI
Analyze historical trade logs to identify performance patterns, profit leakage, and strategy expectancy.
$20
Trading Performance Forensics AI
Example session with this skill installed
Act as my Trading Performance Forensics AI. Analyze this trade history from my London session Breakout strategy: [CSV data with 50 trades including entry/exit, P&L, and duration].
- Read your context and instructions
- Compiled the trading performance forensics
I have analyzed 50 completed trades from the London session Breakout strategy (Date Range: 2023-10-01 to 2023-11-15). The data quality is high with no missing timestamps, though slippage data was not provided and is excluded from leakage calculations.
Trading Performance Forensics Report
Executive summary
- Dataset period: 2023-10-01 to 2023-11-15
- Number of completed trades: 50
- Net P&L: +$1,250.00
- Win rate: 44% (22 Wins, 28 Losses)
- Profit factor: 1.68
- Main finding: High expectancy during the first 90 minutes of the London open, significant performance decay after 10:30 AM GMT.
- Data limitations: Fees and commissions not provided; slippage unknown.
Overall performance
| Metric | Value | Formula/Notes |
|---|---|---|
| Gross Profit | $3,100.00 | Sum of all winning trades |
| Gross Loss | ($1,850.00) | Sum of all losing trades |
| Avg Win | $140.91 | Gross Profit / Win Count |
| Avg Loss | ($66.07) | Gross Loss / Loss Count |
| Expectancy | $25.00 | (0.44 * 140.91) - (0.56 * 66.07) |
Pattern analysis
- By Instrument: GBP/USD (30 trades) showed a 52% win rate; EUR/USD (20 trades) showed a 32% win rate.
- By Time: Trades entered between 08:00-09:30 GMT (25 trades) accounted for 85% of net profits.
- By Direction: Longs (28 trades) outperformed Shorts (22 trades) with a 2.1 profit factor vs 0.9.
Holding-period analysis
- Winning Trades: Median duration 45 minutes.
- Losing Trades: Median duration 120 minutes.
Observation: Losing trades are held nearly 3x longer than winners, suggesting a potential delay in stop-loss execution or "hope" trading.
Actionable review plan
- Reduce Scope: Test restricting entries to the 08:00-09:30 GMT window for 20 trades.
- Directional Bias: Review short-side criteria on EUR/USD to identify why win rates fall below 35%.
- Exit Audit: Review the 5 longest-held losing trades to see if stop-loss rules were manually overridden.
Next steps
- Provide commission/fee data to calculate true net P&L.
- Upload trade logs for the New York session to compare cross-session performance.
- Segment by position size to check for risk-scaling consistency.
Connects securely to your tools. The creator never sees your data.
About this skill
Stop guessing why your P&L is stalling. This skill acts as a forensic analyst for your trading history, digging through raw CSV or Excel logs to identify where you are actually making money and where profit is leaking away. It moves beyond basic win rates to analyze holding periods, session-based performance, and risk consistency.
What it does
- Data validation cleans raw trade logs, handles partial fills, and identifies missing cost data.
- Baseline metrics calculates net P&L, profit factor, expectancy, and win rates using verified formulas.
- Pattern segmentation groups performance by instrument, direction, day of week, and specific trading sessions.
- Holding analysis compares the duration of winning versus losing trades to identify exit behavior patterns.
- Leakage detection quantifies the impact of commissions, slippage, and oversized losses relative to averages.
- Risk auditing evaluates drawdown periods, win/loss streaks, and reward-to-risk ratio consistency.
How it works
- Import trade history via CSV, Excel, or markdown table with entry/exit timestamps and prices.
- Audit data quality to define the scope, timezone, and treatment of breakeven trades.
- Analyze the dataset to generate a structured forensics report with segmented performance metrics.
- Review evidence-based hypotheses for strategy adjustments based on the identified patterns.
Frameworks & tools
This skill is framework-agnostic and processes standard financial data formats including CSV, XLSX, and JSON. It uses mathematical formulas for expectancy and profit factor calculation.
Why this beats prompting it yourself
Generic prompts often hallucinate missing trade data or fail to distinguish between individual fills and completed positions. This skill follows a strict forensic workflow that validates data integrity and requires sample size thresholds before flagging a pattern as significant.
Use cases
- Identify which specific trading sessions or days are yielding negative expectancy.
- Quantify the cost of holding losing trades longer than winners.
- Determine if specific instruments or setups are responsible for the majority of drawdowns.
- Audit execution slippage and fee impact on net profitability.
Known limitations
Analysis is limited by the quality of user-provided logs. It cannot account for market liquidity or psychological factors not captured in the data. Small sample sizes may lead to statistically insignificant conclusions.
Requires trade history with timestamps and prices. Cannot predict future results. Statistical significance depends on sample size. Does not account for unrecorded psychological factors.
How to use
- 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
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 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 16 days ago
Permissions
No special permissions declared or detected