Statistical Analysis Advisor

Recommends appropriate statistical methods (T-test vs ANOVA, etc.) based.

This skill appears coherent and self-contained: it recommends statistical tests, documents assumptions, and includes a local Python script and reference guides. Before installing or executing: 1) Inspect scripts/main.py yourself (or run python -m py_compile scripts/main.py) to confirm there are no unexpected network calls or subprocess usage; the provided excerpts show none but you should verify the whole file. 2) Run the script locally with non-sensitive, synthetic data to confirm behavior. 3) Do not feed secrets or production datasets unless you review how inputs/outputs are handled. 4) Treat the recommendations as advisory — the SKILL.md itself warns that human verification is required for research/publication use.

Type: OpenClaw Skill Name: statistical-analysis-advisor Version: 1.0.0 The statistical-analysis-advisor skill is a legitimate tool designed to recommend statistical methods and perform power analysis. The core logic in scripts/main.py uses standard mathematical formulas and decision trees to provide recommendations based on user-provided data characteristics. There is no evidence of data exfiltration, malicious execution, or unauthorized system access; the script does not even utilize high-risk libraries like os, subprocess, or requests. The documentation in SKILL.md and the references/ directory is purely educational and aligns perfectly with the stated purpose.