Afrexai Data Analyst

What This Skill Does

AfrexAI Data Analyst is a rigorous, framework-driven agent skill designed to transform raw data into actionable business intelligence. Unlike standard chatbots that simply provide summaries, this skill utilizes the DICE framework (Define, Investigate, Communicate, Evaluate) to ensure every piece of analysis serves a specific decision. It guides users through systematic data profiling, hypothesis testing, and root-cause analysis, ensuring that outcomes are statistically sound and business-relevant. It provides a structured workflow for data cleaning, anomaly detection, and cross-referencing multiple data sources, making it an essential tool for turning complex datasets into executive-level insights.

Installation

To integrate this skill into your environment, use the OpenClaw CLI or interface:

clawhub install openclaw/skills/skills/1kalin/afrexai-data-analyst

Use Cases

• Root Cause Analysis: Investigating sudden drops in conversion metrics or revenue fluctuations.
• Sales Strategy Validation: Testing hypotheses about market segment performance or pricing changes.
• Operational Efficiency: Cleaning and analyzing internal logs or spreadsheet data to find bottlenecks in customer support or supply chains.
• Strategic Reporting: Synthesizing complex data from disparate sources like Postgres databases and CSV reports into coherent executive summaries.

Example Prompts

1. "I need to investigate why our Q3 churn rate in the enterprise segment increased by 5%. Please initiate the DICE framework for this analysis and profile the sales and support datasets."
2. "Profile the attached 'user_activity.csv' file and check for data quality issues like missing values or duplicates before we start the investigation phase."
3. "Based on our current marketing spend spreadsheet and the SQL sales data, suggest a hypothesis for why our CAC has spiked in the European market this month."

Tips & Limitations

• Follow the Framework: The skill performs best when you strictly follow the Define phase; the clearer your business question, the more accurate the analysis.
• Data Privacy: Be mindful of sensitive information. Always scrub personally identifiable information (PII) before feeding raw database exports into the analysis loop.
• Complexity: This skill is designed for analytical tasks and is not meant to replace your primary ETL pipeline; it is an investigative layer on top of cleaned or raw source data.
• Verification: Always verify SQL query results, especially when performing complex joins, as database structures can vary significantly between environments.