Outlier Detection & Handling

Use outlier detection handler for data analysis workflows that need structured execution, explicit assumptions, and clear output boundaries.

This skill appears coherent and limited to local outlier analysis. Before installing/running: (1) run it in a sandboxed environment or isolated workspace if you will process sensitive data; (2) verify Python 3.10+ and install requirements.txt in a virtualenv to avoid dependency conflicts; (3) validate and restrict input file paths (avoid symlinks or unexpected directories) to prevent accidental data exposure; (4) review the small scripts/main.py if you need absolute assurance (the code is short and readable); (5) for large datasets, test performance and memory use first. No credentials or network access are required by this skill.

Type: OpenClaw Skill Name: outlier-detection-handler Version: 1.0.1 The skill bundle is a standard implementation for statistical outlier detection using Z-score, IQR, and Grubbs' test. The Python script (scripts/main.py) uses legitimate libraries (numpy, scipy) and follows safe coding practices for data processing, while the SKILL.md provides clear, non-malicious instructions for the AI agent without any signs of prompt injection or unauthorized data access.