csv-analyzer ยท diff
v1.0 to v1.0
1 added, 1 removed. Audit A to A.
What this update changes in the checks
No check changed: every fixed check gives the same answer before and after this update.
A tripped check is a label, not a verdict: read the matched text and the lines below. This week's updates that newly trip a check.
---
name: csv-analyzer
description: Analyze CSV files to generate summary statistics, detect data quality issues, and produce formatted reports.
license: MIT
compatibility: Python 3.10+
metadata:
category: data-processing
version: "1.0"
- allowed-tools: read_file write_file virtual_bash
+ allowed-tools: read_file write_file bash
---
# CSV Analyzer
Analyze CSV data files to produce summary statistics and quality reports.
## When to Use
Activate this skill when a user provides a CSV file and wants:
- Summary statistics (row count, column types, missing values)
- Data quality analysis (duplicates, outliers, formatting issues)
- A formatted report
## Instructions
1. Read the CSV file using the `read_file` tool
2. Run the analysis script at `/skills/csv-analyzer/scripts/analyze.py`
3. Review the output and present findings to the user
## Analysis Steps
### Step 1: Load Data
```bash
python3 /skills/csv-analyzer/scripts/analyze.py /workspace/data.csv
```
### Step 2: Review Output
The script outputs a JSON report with:
- `row_count`: Total number of rows
- `column_count`: Number of columns
- `columns`: Array of column metadata (name, type, null_count, unique_count)
- `quality_issues`: Array of detected problems
### Step 3: Present Results
Format the results as a readable summary table for the user.
## References
See `/skills/csv-analyzer/references/` for additional documentation on CSV analysis best practices.