Protect sensitive information before sharing with AI systems. PII, credentials, and confidential data - redacted in milliseconds.
Edit your redaction configuration directly as JSON
Toggle detection for specific data types:
"patterns": {
"email": { "enabled": true, "strategy": "token" },
"phone": { "enabled": false }
}
john@email.com becomes [EMAIL_1]
john@email.com becomes **************
john@email.com becomes xxxx@xxxxx.xxx
Create patterns for your specific data formats:
"patterns": {
"custom": [
{
"name": "ticketId",
"regex": "TICKET-\\d{5}",
"strategy": "token",
"sampleValue": "TICKET-12345"
}
]
}
name What to call this patternregex Pattern to match (use \\\\ for backslashes)strategy How to redact: token, mask, or formatPreservingsampleValue Example for testing (optional)Add specific names or terms to always redact:
"customEntities": {
"companyNames": ["Acme Corp", "My Company Inc"],
"customerNames": ["John Doe", "Jane Smith"]
}
All built-in patterns have test samples to ensure accuracy:
Each pattern includes 5 test samples covering common cases and edge cases.
Quality Score Breakdown:
Test patterns and preview redaction strategies in real-time
Generate regex patterns from sample data. Mark the specific text you want to match across multiple samples.
1. Paste your sample below 2. Select the text you want to match 3. Click Mark Selection
Help improve Data Redactor by submitting useful patterns to the community library.
Test patterns against curated sample data to ensure accuracy and identify edge cases.
Browse and use patterns submitted by the community. Vote on patterns to help others find the best ones.
MONGODB_URI environment variable and run bun start.
View pattern issues reported by users. Vote on issues and help improve pattern accuracy.
MONGODB_URI environment variable and run bun start.