- Home
- Skills
- Agents & Orchestration
- newbee doc
newbee doc
Professional CSV data quality auditor for identifying missing values, duplicates, and type in
Free
Works with the AI Agents you already use
newbee doc
Example session with this skill installed
Audit this CSV sample
id,date,value,note
1,2023-01-01,10.5,Valid
2,2023-01-02,N/A,Valid
3,2023-01-02,12.0,Valid
1,2023-01-01,10.5,Valid
4,2023-01-04,ERROR,None
- Read your context and instructions
- Compiled the newbee doc
I have completed the audit of the provided CSV sample (5 rows, 4 columns).
CSV Data Quality Audit Report
Audit Basis: Partial sample (5 rows)
1. Summary Table
| Category | Status | Notes |
|---|---|---|
| Missing Values | PASS | No empty cells detected. |
| Duplicate Rows | FAIL | 1 duplicate row found. |
| Type Consistency | WARN | Mixed types in 'value' column. |
| Null Encodings | FAIL | Found 'N/A' and 'None'. |
| Constant Columns | WARN | 'note' column has low variance. |
2. Detailed Findings
- Duplicate Rows: Row 4 is a duplicate of Row 1.
- Type Mismatch (value): Column 'value' is primarily numeric, but contains 'ERROR' (Row 5).
- Null Encodings:
- Column 'value': 'N/A' detected (Row 2).
- Column 'note': 'None' detected (Row 5).
Constant/Low Variance: Column 'note' contains 'Valid' in 80% of rows; check if this column provides unique information.
3. Incomplete Checks
- Large Scale Patterns: Unable to verify patterns across full dataset as only 5 rows were provided.
4. Privacy Notice
- No PII (names, emails, phones) detected in this sample.
Scope Reminder: This is an audit report only. No data has been cleaned or modified.
Next steps
- Remove the duplicate entry at Row 4.
- Investigate the 'ERROR' string in the 'value' column at Row 5.
- Replace 'N/A' and 'None' strings with standard null values during your import process.
Connects securely to your tools. The creator never sees your data.
About this skill
Dirty data ruins analysis. This skill audits CSV files to identify structural and content issues before you waste time on broken pipelines. It systematically scans for missing values, type mismatches, and duplicate records without modifying the source.
What it does
- Missing Value Detection scans every column to report null counts and percentages.
- Duplicate Identification flags exact row duplicates within the provided dataset.
- Type Consistency Check infers column types and isolates cells that deviate from the majority format.
- Null Encoding Analysis finds stealthy missing values like
N/A,?, orNULLthat bypass standard loaders. - Constant Column Flagging identifies columns with zero variance that offer no analytical value.
How it works
- Analyze Coverage by checking if the input is a full file or a sample snippet to set audit scope.
- Execute Scans across five quality dimensions defined in the internal ruleset.
- Log Anomalies by specific column and row index for easy manual verification.
- Generate Report using a standardized Markdown template with PASS/WARN/FAIL statuses.
Frameworks & tools
This skill is framework-agnostic and processes raw CSV text or file uploads. It uses internal logic to simulate pandas-like profiling without requiring an execution environment.
Why this beats prompting it yourself
Generic prompts often miss non-standard null encodings (like nan vs None) or fail to check for constant columns. This skill follows a rigid audit checklist and output schema, ensuring no quality dimension is skipped and the report remains scannable.
Use cases
- Auditing raw data exports before importing them into a production SQL database.
- Verifying third-party CSV samples during the vendor data evaluation phase.
- Checking for PII leaks in datasets before sharing them with external teams.
- Validating automated report exports for consistency and logic errors.
Known limitations
Does not perform data cleaning or transformations. Limited to CSV format only. Samples under 3 rows will return incomplete audit results.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Agents & Orchestration.
Cannot clean or transform data. Limited to CSV files. Samples under 3 rows trigger incomplete status. Does not support Excel or JSON.
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 buying a skill.
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 to use
No account required to browse
Trust & safety
Security scan
Security scan passed today
- Free with an Agensi account
Permissions
No special permissions declared or detected