DOL Enforcement Rollup — Employer Wage-Theft Violator API

Roll up US Department of Labor enforcement actions by employer. Aggregates Wage and Hour Division (WHD) cases — back wages, civil money penalties, employees affected, repeat/willful flags — into per-employer summaries with repeat-violator flags, or returns one normalized row per case.

Who it's for

• Procurement & vendor risk teams screening suppliers and staffing agencies.
• ESG / labour-rights analysts and journalists tracking wage theft.
• Union research and worker-advocacy organisations.
• Diligence teams pricing labour liability into an acquisition.
• AI agents needing a "check this employer's DOL enforcement history" tool.

It never tells you an employer is clean when it did not get an answer

This is a screening product, so the expensive direction of wrong is the negative: an employer reported as having no enforcement history because a request failed. Three guarantees:

1. The DOL endpoint is health-checked live before any result is returned — reachable, still serving the columns the mapper reads, still applying the filter (a negative control query that must match nothing), no unknown value in the repeat/willful vocabulary, and a positive canary that must return rows.
2. If no source answers, the run FAILS and bills nothing. It does not return an empty dataset, which would read as a clean record.
3. Every row says how complete it is — enforcement_history_complete, partial_results, sources_failed, result_truncated. When the answer is incomplete, a negative repeat_violator verdict is emitted as null, never false, and repeat_violator_basis says why.

Zero rows with every source healthy is a real answer — a genuine clean result — and the log says so explicitly.

Getting a key

Free and instant at https://dataportal.dol.gov/registration. Pass it as apiKey, or set the DOL_API_KEY environment variable. Without one the run fails with an actionable message rather than returning an empty dataset.

Example input

{
  "mode": "rollup",
  "state": "CA",
  "sinceDate": "2023-01-01",
  "maxResults": 500
}

Look up one employer's full history:

{ "mode": "cases", "employer": "tyson foods", "maxResults": 1000 }

Output fields

Every field below is declared in the dataset schema. Fields are nullable, and null means DOL does not publish it or the value was not determinable — never "we checked and it is zero".

Employer identity (both modes)

Field	What it is
employer_name	Employer name, whitespace-normalised.
employer_name_raw	The name exactly as DOL publishes it.
legal_name	Registered legal name, where DOL publishes it separately from the trade name.
agency	WHD, EBSA, OFCCP, or MULTIPLE on a rollup spanning agencies.
agencies	Every agency contributing to a rollup row.
source_agency	Which DOL agency produced a per-case row.
naics · naics_description	Industry code and its description.
city · state · zip · street_address	Establishment location.
source_url	DOL enforcement-data reference page.
retrieved_at	When this run pulled the data.
data_load_date	Date DOL loaded the record into the published extract.

Per-case rows (mode: "cases")

Field	What it is
case_id	DOL case identifier.
back_wages	Back wages agreed to be paid on the case.
back_wages_minimum_wage · back_wages_overtime	The FLSA split, where DOL breaks it out.
civil_penalty	Sum of the per-statute *_cmp_assd_amt penalty columns.
civil_penalty_case_total	The case-level penalty DOL publishes in cmp_assd.
civil_penalty_matches_case_total	true when those two reconcile. false means DOL assessed a penalty not broken out by statute, so civil_penalty understates it. null when DOL published no case total.
employees_affected	Employees found in violation (ee_violtd_cnt).
employees_paid	Employees the employer agreed to pay (ee_atp_cnt) — a different number.
minors_employed_illegally	Minors employed in violation of FLSA child-labour provisions.
violation_count	Violations recorded on the case.
violation_type	The single largest statute violated.
violation_types_all	Every statute with a recorded violation, semicolon-separated — a multi-statute case is not one label.
repeat_violator_flag	true when DOL marks the case repeat or willful.
repeat_violator_code	The raw code: R, W, RW or N/A.
finding_start_date · finding_end_date	Investigation period.

Per-employer rollup rows (mode: "rollup", the default)

Field	What it is
total_cases	Cases rolled into this employer. A floor, not a count, when total_cases_is_complete is false.
total_cases_is_complete	false when the fetch was capped or a source failed.
total_back_wages · total_civil_penalties · total_employees · total_violations	Sums across the employer's cases.
total_civil_penalties_case_reported	Sum of DOL's case-level penalty column, for comparison with total_civil_penalties.
total_employees_paid · total_minors_employed_illegally	Sums of the corresponding per-case fields.
total_back_wages_minimum_wage · total_back_wages_overtime	The FLSA split, summed.
largest_single_case_back_wages	Separates one large case from a long tail of small ones.
first_finding · last_finding	Earliest start and latest end across the employer's cases.
repeat_violator	true / false / null when not determinable.
repeat_violator_basis	case_count_and_whd_flag, whd_repeat_or_willful_flag, case_count, no_indicator_found, or not_determinable_incomplete_data. Read this before acting on a negative.
whd_repeat_or_willful_cases	How many of the employer's cases carry the DOL designation.
distinct_source_names · source_names_merged	Grouping normalises corporate suffixes, so "ABC Foods Inc" and "ABC Foods LLC" collapse together. These let you audit any merge.

Run integrity — on every row

Field	What it is
enforcement_history_complete	The field to read before acting on a negative. true only when every requested source answered and nothing was truncated.
partial_results	true when at least one requested source did not answer. Computed from the actual outcomes every run, never a hard-coded false.
sources_requested · sources_ok · sources_failed · sources_failed_detail	Exactly which DOL sources answered, which did not, and why.
result_truncated · truncated_sources	true when a source returned exactly maxResults cases, so more exist than were counted.
run_mode	rollup or cases.

Use as an MCP tool

Exposed to AI agents via mcp.apify.com as a "check an employer's DOL enforcement history" tool. The per-source status fields matter here: an agent can distinguish a verified clean record from an unavailable source instead of reporting the second as the first.

FAQ

Is this a background check? No. It is a screening tool over DOL's own published enforcement extract, not a determination about any employer. Confirm against the DOL case record before acting.

Why is repeat_violator sometimes null? Because the fetch was capped by maxResults, or a source did not answer. total_cases is then a floor, and a "not a repeat violator" verdict would be a claim the data cannot support. Raise maxResults for a determinate answer.

Why did I get zero rows? If sources_failed is null on the log line, every source answered and genuinely had nothing — a real clean result for those filters. Otherwise the run fails loudly instead.

Which agencies are covered? WHD by default. EBSA (ERISA Form 5500 penalties) is opt-in via sources. OFCCP is scaffolded but not enabled — no keyless DOL-portal endpoint has been verified for it.

How far back does it go? WHD publishes concluded compliance actions from FY2005 onward.

Source

US Department of Labor Open Data API v4 (apiprod.dol.gov/v4/get), agency WHD, endpoint enforcement. Free key required. DOL caps a single response at 6 MB, so this actor paginates in verified-safe pages and shrinks automatically rather than failing.

Pricing (Pay Per Result)

Billed per row returned, with graduated discounts on paid Apify plans. A run that cannot reach DOL fails instead of billing.

Related actors

• Bank Enforcement Tracker — OCC / FDIC enforcement actions by institution.
• License Verifier — professional and occupational licence status, 19 boards.
• KYB Company Verifier · US Business Entity Search — entity registration and standing, to attach an enforcement history to a verified legal entity.