# Csv Analysis

> Analyze CSV files and generate comprehensive statistical reports with data profiling and quality checks

- **Type:** Skill
- **Install:** `agentstack add skill-codebytes-agent-skills-csv-analysis`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [codebytes](https://agentstack.voostack.com/s/codebytes)
- **Installs:** 0
- **Category:** [Data & Analytics](https://agentstack.voostack.com/c/data-and-analytics)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [codebytes](https://github.com/codebytes)
- **Source:** https://github.com/codebytes/agent-skills/tree/main/plugins/document-tools/skills/csv-analysis

## Install

```sh
agentstack add skill-codebytes-agent-skills-csv-analysis
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

## Instructions

When asked to analyze a CSV file, follow this workflow:

### Step 1: Read and Profile
- Use the `view` tool to read the first 50 lines of the CSV
- Identify the delimiter (comma, tab, semicolon, pipe)
- Count total rows and columns
- Infer column data types (string, integer, float, date, boolean)

### Step 2: Compute Statistics
Run a Python script to compute per-column statistics:

```python
import csv
import statistics
from collections import Counter

# Read and analyze the CSV
# Compute: count, nulls, unique values, min, max, mean, median, std dev
```

### Step 3: Quality Assessment
Check for:
- Missing or null values (empty strings, "NA", "null", "N/A")
- Duplicate rows
- Inconsistent formatting (mixed date formats, case inconsistency)
- Potential outliers (values beyond 3 standard deviations)

### Step 4: Generate Report
Create a markdown report with:
- **Overview**: File name, row count, column count
- **Schema table**: Column name, type, non-null count, unique count
- **Statistics table**: Min, max, mean, median, std dev for numeric columns
- **Quality issues**: List of findings with severity (info/warning/error)
- **Key findings**: Top 3-5 insights from the data

## Output Format

The report should be a well-formatted markdown document suitable for inclusion in project documentation. Use tables for structured data and bullet points for findings.

## Error Handling

- If the file is not valid CSV, report the issue and suggest the correct format
- If the file is too large (>100MB), sample the first 10,000 rows and note the sampling
- If encoding errors occur, try UTF-8, Latin-1, and CP1252 in order

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [codebytes](https://github.com/codebytes)
- **Source:** [codebytes/agent-skills](https://github.com/codebytes/agent-skills)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-codebytes-agent-skills-csv-analysis
- Seller: https://agentstack.voostack.com/s/codebytes
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.