- Home
- Skills
- Data & Databases
- CSV Data Quality Auditor for Cleaner Datasets
CSV Data Quality Auditor for Cleaner Datasets
Audit CSV files for missing values, duplicates, and type mismatches with a detailed Markdown quality report.
Free
Works with the AI Agents you already use
CSV Data Quality Auditor for Cleaner Datasets
Example session with this skill installed
Audit data/user_leads.csv. The primary key is 'email'. Check if 'signup_date' is a valid date and 'age' is an integer.
- Read your context and instructions
- Compiled the csv data quality
- Generated the data export
I have completed the audit of data/user_leads.csv in read-only mode. No data has been modified.
CSV Data Quality Audit Report
Status: REVIEW REQUIRED
Summary
The audit detected 12 confirmed issues across 1,000 rows. Structural integrity is high, but type mismatches were found in the age column.
Confirmed Issues
| Issue Type | Column | Count | Impact |
|---|---|---|---|
| Type Mismatch | age | 8 | Critical |
| Missing Values | signup_date | 4 | High |
| Duplicate ID | 2 | Medium |
Missing Values by Column
- signup_date: 4 missing values found (Rows 102, 445, 890, 912).
- email: 0 missing values.
- age: 0 missing values.
Type / Encoding / Column Mismatch Examples
- Column
age: Expectedinteger. Found strings:"25yo","unknown","N/A"in rows 12, 45, and 67. - Column
signup_date: All present values match the expected date format.
Duplicate Examples
- email:
test@example.comappears in rows 14 and 502.
Pending Confirmations
- Please confirm if
"N/A"in theagecolumn should be treated as a missing value or a type error. - Verify if the encoding
UTF-8is correct for your source system.
Limitations
This report is auxiliary and does not certify data for financial or legal use. Always perform a manual spot check before ingestion.
Next steps
- Clean the 8 rows in
agecontaining non-numeric strings. - Resolve the duplicate
emailentries in rows 14 and 502. - Re-run the audit after cleaning to verify the fix.
user_leads.csv
CSV · data export
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Data analysis often fails because of silent failures in raw CSV files. This skill audits CSVs for structural defects, type mismatches, and missing values before you start your pipeline.
What it does
- Structural validation checks for header consistency, encoding anomalies, and column mismatches.
- Null detection identifies missing values across all columns based on provided or common tokens.
- Duplicate auditing finds and flags duplicate rows or ID-based collisions.
- Type verification compares actual data against an expected schema for integers, floats, dates, and booleans.
- Issue reporting generates a reviewable Markdown checklist with specific examples of every error found.
How it works
- Source connection identifies the CSV file path and ensures read-only access.
- Execution runs the audit script using provided ID columns and schema definitions.
- Manual fallback applies internal rulesets if automated scripts are restricted by the environment.
- Report generation produces a formatted quality summary including confirmed issues and pending confirmations.
Frameworks & tools
The skill utilizes Python scripts for data parsing and follows standard Markdown reporting templates. It operates on flat CSV files and relies on local or provided file paths.
Why this beats prompting it yourself
Generic prompts often miss edge cases like encoding errors or subtle type mismatches in large files. This skill follows a rigid audit protocol that ensures no column is skipped and every issue is backed by concrete examples.
Use cases
- Audit raw datasets before training machine learning models.
- Validate CSV exports from third-party tools before database ingestion.
- Identify corrupted rows in large files that cause script crashes.
- Verify data types and null counts for periodic reporting tasks.
Known limitations
Does not perform destructive cleaning, multi-file ETL, or statistical analysis. Requires manual de-identification of sensitive data before auditing.
Does not modify or clean data. Limited to single CSV files. Cannot infer business logic or legal compliance. Large files may require user confirmation before processing.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed today
- Free with an Agensi account