A 12-dimensional technical feature attribution engine that identifies the market microstructure conditions separating winning trades from losing ones.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install qc-deep-feature-forensics
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install qc-deep-feature-forensics using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is QC Deep Feature Forensics?
QC Deep Feature Forensics is a sophisticated technical analysis tool designed for traders and developers who need to look beyond simple profit and loss metrics. By utilizing the Openclaw Skills framework, this engine performs a deep forensic audit of trade entries, comparing the microstructure of winning trades against losing ones across 12 distinct dimensions. It identifies exactly what market conditions—such as RSI levels, volume surges, or Bollinger Band positioning—were present at the moment of entry to determine what drives success.
This skill is essential for anyone looking to refine their trading edge through statistical evidence. It reconstructs trade history into closed pairs and integrates with financial data providers to build a comprehensive feature matrix. By using this tool within the Openclaw Skills ecosystem, users can move from intuitive guessing to data-driven strategy optimization, ensuring their entry conditions are aligned with high-probability market environments.
QC Deep Feature Forensics Use Cases
- Identifying specific technical indicators that correlate with winning trade entries.
- Performing deep-dive feature attribution on historical backtests to improve strategy performance.
- Testing hypothetical market filters using what-if simulations to see their impact on total portfolio ROI.
- Diagnosing why certain trading setups consistently fail by analyzing underlying market microstructure.
- Automating the generation of an ideal entry profile for algorithmic trading development.
How QC Deep Feature Forensics Works
- Reconstructs raw order data into consolidated closed trade pairs by grouping buy and sell executions.
- Automatically downloads historical daily OHLCV data for all relevant tickers using the Yahoo Finance API.
- Pre-computes 12 specific technical factors for the entire ticker history, including volatility expansion and moving average deviations.
- Extracts a 12-dimensional feature vector for every individual trade at its specific entry timestamp.
- Executes a statistical dual-sample comparison between winning and losing trades to identify significant performance drivers.
- Performs a what-if simulation to test how various filters would have affected the total net P&L and portfolio ROI.
QC Deep Feature Forensics Setup
To deploy this tool as part of your Openclaw Skills library, ensure you have Python 3 and the necessary data science libraries installed.
pip3 install pandas numpy yfinance
Once dependencies are installed, you can execute the forensics engine by providing your trade history file:
python3 deep_forensics.py <orders.csv>
QC Deep Feature Forensics Data Schema & Taxonomy
The skill organizes its output into structured diagnostic files and maintains a local cache to optimize subsequent runs within the Openclaw Skills environment:
| File Name | Content | Purpose |
|---|---|---|
<name>_features.csv |
Full Feature Matrix | Contains all 12 calculated technical factors for every trade entry. |
feature_diagnosis.md |
Markdown Report | The final diagnostic summary including winner/loser comparisons and what-if results. |
yfinance_cache/ |
OHLCV Data Cache | Local CSV storage of ticker history to reduce API calls and improve speed. |
Primary Feature Factors Analyzed:
volume_surge: Current volume relative to the 10-day average.rsi_14: Relative Strength Index value at entry.bb_position: Relative position within the Bollinger Bands (0 to 1).gap_pct: The percentage gap between the open and the previous close.volatility_expansion: Current intraday range relative to the 5-day ATR.
QC Deep Feature Forensics Advanced Features
- Intelligent What-If simulation engine that classifies filters as Shield, Toxic, or Marginal based on total ROI impact.
- Automated local caching system that checks for date-range coverage to prevent redundant data downloads.
- Statistical percentile profiling to define the ideal entry environment envelope for winning trades.
- Combined filter stacking to measure the aggregate performance impact of multiple Shield filters.
- Multi-dimensional attribution that goes beyond simple indicators to include gap analysis and consecutive day trends.
SKILL.md
Loading
Related Openclaw Skills
IMAP IDLE Email Notifier
A real-time email monitoring skill that uses IMAP IDLE to push instant notifications to Feishu.
Cognitive Memory System
An intelligent multi-store memory system for AI agents featuring human-like encoding, reflection cycles, and decay-based forgetting.
Fund Investment Advisory Assistant (fin-advisor)
A specialized AI investment assistant that delivers professional fund analysis, market comparisons, and data-driven decision support.
Brave API Search
A powerful suite of tools for real-time web search, intelligent query suggestions, and AI-driven answers grounded in live web data.
Backtest Poller
An intelligent background daemon for Openclaw Skills that automates QuantConnect backtest monitoring, equity tracking, and forensic diagnosis.
QC Backtest Master
A streamlined pipeline for automating QuantConnect backtest lifecycles from local strategy files to detailed cloud result downloads.
Featured*
