Privacy Scrubber.

What this tool helps with

Sensitive data leaks rarely happen because someone intended to publish a secret. They usually happen because a log snippet, API response, CSV export, XML payload, or support transcript was copied too quickly into a ticket, chat, bug report, or shared document.

This privacy scrubber is designed for that exact moment. Paste the content you want to share, mask the values that should not leave your machine, then copy the cleaned version onward. The first version is especially good for JSON, but it can also scrub CSV, XML, pasted logs, and other text-based formats.

When to use it

• before posting logs into Slack, Teams, Discord, or email
• before filing a support ticket with exported account data
• before sharing API payloads or webhook samples with clients
• before sending CSV rows or XML snippets during debugging
• before pasting copied output into documentation or public issues

What it can scrub today

The current version looks for several common categories of sensitive data:

• email addresses
• phone numbers
• IPv4 addresses
• UUIDs and GUIDs
• JWTs, API keys, bearer-token style strings, and other secret-looking values
• JSON fields with sensitive names such as token, password, secret, apiKey, email, or phone

If the input is valid JSON, the tool parses it and rewrites values more carefully so the output stays valid JSON. For CSV, XML, and other text formats, it applies pattern-based scrubbing directly to the pasted text.

How to get the best result

1. Paste the original content into the input box.
2. Leave the format on Auto detect unless you know you want to force JSON, CSV, XML, or plain text mode.
3. Keep the default scrubbing categories enabled unless you need a narrower pass.
4. Click Scrub Sensitive Data.
5. Review the output once before sharing it.

JSON, CSV, XML, and other formats

Yes, the tool can work beyond JSON.

• JSON gets the strongest handling because it can be parsed and rewritten structurally.
• CSV and TSV are scrubbed as text, which is usually enough for emails, tokens, phone numbers, UUIDs, and IPs inside rows and cells.
• XML and HTML-like payloads are also scrubbed as text, including sensitive-looking attribute values and inline content.
• Other formats such as logs, copied terminal output, config fragments, and mixed snippets still benefit from the generic text scrubber.

That means v1 is already useful across multiple formats, while leaving room for deeper format-specific logic later.

What this tool does not do

This first version does not try to understand every schema or every proprietary token format.

It does not:

• guarantee that every possible secret will be detected
• validate whether a CSV or XML document remains semantically correct after redaction
• detect all international phone-number edge cases perfectly
• replace human review for high-risk data sharing

That limitation matters. A scrubber should reduce obvious leakage fast, but users should still review the result before posting anything externally.

A practical workflow

For support and debugging work, a reliable flow is:

1. copy the raw payload, export, or log block
2. scrub it locally in the browser
3. review the redacted output for anything still too specific
4. share the cleaned version in the ticket, issue, or chat
5. keep the original private source out of the conversation

That gives you a repeatable, low-friction privacy step instead of relying on manual search-and-replace every time.