retail-trade-report-generator

Generates a consolidated weekly retail trade report by processing 12 Excel sales files, mapping stores to regions, calculating ADA metrics, WoW comparisons,...

This package looks coherent and not malicious, but take these practical steps before running it: 1) Review the Python code yourself (or have a developer do so) to confirm the column/row indices and sheet names match your real Excel exports — the script uses hard-coded row ranges and sheet names which may need adjustment. 2) Run it in a controlled environment (local VM or container) and install dependencies from official sources (pip install pandas openpyxl). 3) Provide only the intended input files and the store_mapping.csv; the skill reads only files in the specified input directory and writes an output Excel. 4) Test with sample data to verify results and edge-case handling (e.g., missing sheets, date parsing). If you need the skill to run autonomously on a platform, ensure the runtime includes the required Python packages or add an install step that pins trusted package versions.

Type: OpenClaw Skill Name: retail-trade-report-generator Version: 1.0.0 The skill bundle is classified as benign. The `SKILL.md` documentation clearly outlines the purpose and functionality, providing instructions for an AI agent that are aligned with data processing and report generation without any signs of prompt injection or malicious intent. The Python script `scripts/retail_trade_report_generator.py` uses standard libraries (`pandas`, `openpyxl`, `re`, `os`, `datetime`) for reading specified Excel and CSV files from an input directory and writing a consolidated Excel report to an output directory. There is no evidence of data exfiltration, malicious execution (e.g., `eval`, `subprocess`, network calls), persistence mechanisms, or obfuscation. All file operations are confined to expected user-data paths.