Qc Deep Feature Forensics
v1.0.012-dimensional technical feature attribution engine — compares winner vs loser trade entry conditions using RSI, Bollinger, MACD, volume surge, gap, and more...
⭐ 0· 92·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (12-dimensional feature attribution for winner vs loser trade entries) matches the code and SKILL.md. The script reads an orders CSV, reconstructs closed trades, downloads OHLCV from Yahoo via yfinance, computes indicators, and writes feature and report files. Required binaries (python3, pip3) and the listed pip dependencies are appropriate for the task.
Instruction Scope
Runtime instructions are narrow: install the listed Python packages, run `python3 deep_forensics.py <orders.csv>`. The code only reads the provided orders CSV, writes outputs and a per-ticker cache directory next to the CSV, and makes network calls to Yahoo Finance via yfinance. There are no instructions to read unrelated system files, environment variables, or to post data to unknown endpoints.
Install Mechanism
There is no packaged installer; SKILL.md instructs doing `pip3 install pandas numpy yfinance` and a requirements.txt is included. This is a normal approach for a Python script, but pip installs run arbitrary package code from PyPI — recommend using a virtual environment or isolated environment when installing. No downloads from unknown URLs or archive extraction are present.
Credentials
The skill requests no environment variables, credentials, or config paths. Its network use is limited to yfinance (Yahoo Finance) for historical data, which is consistent with the functionality. No unrelated secrets are requested or accessed.
Persistence & Privilege
The skill does not request always:true and is user-invocable only. It writes a local cache directory (yfinance_cache) and output CSV/markdown next to the orders CSV — typical and proportionate for caching. It does not modify other skills or global agent settings.
Assessment
This skill appears coherent and implements what it claims, but before running: (1) install dependencies inside a virtualenv/container to limit pip risk; (2) review the orders CSV you supply (it will be read and used as the sole input) and ensure it contains only the data you intend to analyze; (3) be aware the script will write a yfinance_cache/ folder beside your orders CSV and output files (feature CSV and feature_diagnosis.md); (4) the script requires internet on first run to fetch Yahoo data — if you need offline runs, pre-populate the cache; (5) if you handle sensitive trading/account data, inspect the full script locally (the included Python file appears to log only basic status messages) and run in an isolated environment. Overall the behavior is proportionate to the stated purpose.Like a lobster shell, security has layers — review code before you run it.
latestvk97dxfaed2ktdeb1h5dc53g1w9832dx5
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🧬 Clawdis
Binspython3, pip3