Skip to content

HR Detectors ​

hr_identifier ​

Detects HR identifiers such as employee IDs, staff numbers, payroll numbers, and timesheet numbers. Requires nearby context labels.

ts
const redactor = createRedactor({
  rules: { hr_identifier: { action: "redact" } },
});

redactor.redact("Employee ID: EMP123456");
// "Employee ID: [HR_IDENTIFIER_1]"
  • ID: hr_identifier
  • Entity type: hr_identifier
  • Context required: Yes (labels: Employee ID, EMP-ID, Staff No, Personnel ID, Worker ID, Payroll No, PAY ID, Timesheet No, Timecard ID, Time-Entry No)
  • Stream supported: Yes (maxMatchLength: 25)
  • Validation: Context label presence

hr_screening ​

Detects HR screening IDs such as background check IDs, drug test IDs, disciplinary action numbers, and incident numbers. Requires nearby context labels.

ts
const redactor = createRedactor({
  rules: { hr_screening: { action: "redact" } },
});

redactor.redact("Background Check ID: BGC1234567");
// "Background Check ID: [HR_SCREENING_1]"
  • ID: hr_screening
  • Entity type: hr_screening
  • Context required: Yes (labels: Background Check ID, BGC ID, Screening ID, Drug Test ID, Urinalysis ID, Disciplinary Action No, Incident No, Warning No, Violation No)
  • Stream supported: Yes (maxMatchLength: 25)
  • Validation: Context label presence, at least one digit required

hr_compensation ​

Detects compensation data including salary amounts and benefit account numbers. Requires nearby context labels. Uses dual patterns with deduplication — one for currency amounts and one for alphanumeric account IDs.

ts
const redactor = createRedactor({
  rules: { hr_compensation: { action: "redact" } },
});

redactor.redact("Salary: $85,000.00");
// "Salary: [HR_COMPENSATION_1]"
  • ID: hr_compensation
  • Entity type: hr_compensation
  • Context required: Yes (labels: Salary, Compensation, Pay, Wage, Earning, Benefits Plan No, Insurance Plan ID, Health-Plan No, 401K Account No, 403B No, IRA No, Retirement Account No, Pension No)
  • Stream supported: Yes (maxMatchLength: 30)
  • Validation: Context label presence, dual pattern matching with deduplication

hr_recruitment ​

Detects recruitment IDs such as application IDs, candidate IDs, resume IDs, performance review IDs, and training certification IDs. Requires nearby context labels.

ts
const redactor = createRedactor({
  rules: { hr_recruitment: { action: "redact" } },
});

redactor.redact("Application ID: APP1234567");
// "Application ID: [HR_RECRUITMENT_1]"
  • ID: hr_recruitment
  • Entity type: hr_recruitment
  • Context required: Yes (labels: Application ID, Candidate ID, Applicant No, Application Ref, Resume ID, CV No, Performance ID, Review ID, Appraisal No, Evaluation ID, Training ID, Certification ID, Cert No, Recruiter Ref, Agency ID)
  • Stream supported: Yes (maxMatchLength: 25)
  • Validation: Context label presence, at least one digit required

Released under the MIT License.