Data Analyst

---
name: data-analyst
description: |
  Analyze datasets with statistical methods. Trigger when the user has data to
  analyze, needs statistics, or wants insights from CSV/JSON data. Phrases:
  "analyze this data", "what are the trends", "run statistics", "correlations".
allowed-tools:
  - Bash(python3 *)
  - Bash(pip install *)
  - Read
  - Write
---

# Data Analyst

Analyze datasets using statistical methods: descriptive stats, correlations,
hypothesis testing, trend analysis, and visualization recommendations.

## Workflow

1. **Ingest** - Read the data file (CSV, JSON, Excel, Parquet)
2. **Profile** - Shape, dtypes, missing values, basic stats
3. **Clean** - Handle nulls, outliers, type coercions
4. **Analyze** - Run requested analysis (stats, correlations, trends)
5. **Report** - Summarize findings with key takeaways

## Example usage

- "Analyze this CSV and tell me what is interesting"
- "What is the correlation between revenue and marketing spend?"
- "Run a hypothesis test: did the new feature improve retention?"
- "Show me monthly trends for the past year"

## Python libraries used

```bash
pip install pandas numpy scipy matplotlib seaborn
```

## Output format

- Summary statistics table
- Key findings (bulleted, ranked by significance)
- Suggested next analyses
- Code snippets for reproducibility

## Requirements

- Python 3.9+
- Data file accessible from the working directory