Code · Made a little simpler

CSV PII Column Detector

Check local CSV columns for whole-cell patterns and recognized header clues. Review heuristic confidence, column positions and counts without exporting header names or values.

Free to useNo account neededProcessed on your device
Your workspaceProcessed on your device

Add this tool to your website

Paste this code into your page to add the tool and a visible ToolMellow credit link.

HTML embed code
Preview embed

CSV PII Column Detector: how to use it

  1. Paste a table or open a local file. Choose its delimiter and confirm whether the first record contains headers.
  2. Select Scan columns. Compare matching cells with each column’s nonempty count and review the confidence labels.
  3. Investigate the flagged columns in your source. Copy or download the aggregate report for review.

Limits and notes

Reads up to 1,000,000 UTF-16 code units, 128 columns, 10,000 data records and 250,000 parsed cells. Checks whole-cell patterns for common email addresses, formatted phones, IP literals and hyphenated US SSN-shaped strings. Name, address and personal identifier indicators come from selected English headers. The scan cannot determine whether a value identifies a person.

How your privacy works →

CSV PII Column Detector: common questions

Does the report expose the values it matched?

The report includes column positions, signal names, counts, confidence and whether a recognized header contributed. It leaves out source values and header names.

Why is a phone-like column only marked Possible signal?

Strong signal requires three matches covering at least 75% of nonempty cells, or an agreeing header and two matches covering at least half. Mixed or smaller samples can still yield a possible match.

Does the scan search notes for names or emails?

Value patterns apply to whole cells. Prose, personal names and addresses need separate review, even when no signal appears.

Guides