{
"openapi": "3.0.1",
"info": {
"title": "MSHA Mine, Quarry & Aggregates Operator Leads",
"description": "Scrape US mine, quarry & aggregates operators from official MSHA data: operator & parent controller, commodity, employees, full address, geo, production & violation/penalty intel. Filter by state, commodity, status & size. B2B leads + monitoring, no browser.",
"version": "0.1",
"x-build-id": "ozqL7gqZbaGEgLqKG"
},
"servers": [
{
"url": "https://api.apify.com/v2"
}
],
"paths": {
"/acts/scrapesage~msha-mine-operations-scraper/run-sync-get-dataset-items": {
"post": {
"operationId": "run-sync-get-dataset-items-scrapesage-msha-mine-operations-scraper",
"x-openai-isConsequential": false,
"summary": "Executes an Actor, waits for its completion, and returns Actor's dataset items in response.",
"tags": [
"Run Actor"
],
"requestBody": {
"required": true,
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/inputSchema"
}
}
}
},
"parameters": [
{
"name": "token",
"in": "query",
"required": true,
"schema": {
"type": "string"
},
"description": "Enter your Apify token here"
}
],
"responses": {
"200": {
"description": "OK"
}
}
}
},
"/acts/scrapesage~msha-mine-operations-scraper/runs": {
"post": {
"operationId": "runs-sync-scrapesage-msha-mine-operations-scraper",
"x-openai-isConsequential": false,
"summary": "Executes an Actor and returns information about the initiated run in response.",
"tags": [
"Run Actor"
],
"requestBody": {
"required": true,
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/inputSchema"
}
}
}
},
"parameters": [
{
"name": "token",
"in": "query",
"required": true,
"schema": {
"type": "string"
},
"description": "Enter your Apify token here"
}
],
"responses": {
"200": {
"description": "OK",
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/runsResponseSchema"
}
}
}
}
}
}
},
"/acts/scrapesage~msha-mine-operations-scraper/run-sync": {
"post": {
"operationId": "run-sync-scrapesage-msha-mine-operations-scraper",
"x-openai-isConsequential": false,
"summary": "Executes an Actor, waits for completion, and returns the OUTPUT from Key-value store in response.",
"tags": [
"Run Actor"
],
"requestBody": {
"required": true,
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/inputSchema"
}
}
}
},
"parameters": [
{
"name": "token",
"in": "query",
"required": true,
"schema": {
"type": "string"
},
"description": "Enter your Apify token here"
}
],
"responses": {
"200": {
"description": "OK"
}
}
}
}
},
"components": {
"schemas": {
"inputSchema": {
"type": "object",
"properties": {
"states": {
"title": "States",
"type": "array",
"description": "Filter by the state the mine is located in — two-letter USPS codes, e.g. TX, WV, PA, CA, KY. Leave empty for all states & territories.",
"items": {
"type": "string"
}
},
"commodityGroups": {
"title": "Commodity groups",
"type": "array",
"description": "Filter by what the operation mines. Sand & Gravel and Stone are the construction-aggregates producers (the largest, most accessible B2B segment); Coal, Metal and Nonmetal cover everything else. Leave empty for all.",
"items": {
"type": "string",
"enum": [
"coal",
"sandGravel",
"stone",
"nonmetal",
"metal"
],
"enumTitles": [
"Coal",
"Sand & Gravel",
"Stone",
"Nonmetal",
"Metal"
]
}
},
"coalOrMetal": {
"title": "Coal or Metal/Nonmetal",
"enum": [
"any",
"coal",
"metalNonmetal"
],
"type": "string",
"description": "Quick high-level filter. Use the more specific Commodity groups above when you need it.",
"default": "any"
},
"counties": {
"title": "Counties",
"type": "array",
"description": "Filter by county name (case-insensitive), e.g. Harris, Clearfield, Elko.",
"items": {
"type": "string"
}
},
"mineStatuses": {
"title": "Mine statuses",
"type": "array",
"description": "Operational status. Defaults to Active — operating businesses that make the best leads. Add New Mine for brand-new, high-intent operations, or others to widen the net.",
"items": {
"type": "string",
"enum": [
"Active",
"New Mine",
"Intermittent",
"NonProducing",
"Temporarily Idled",
"Abandoned",
"Abandoned and Sealed"
],
"enumTitles": [
"Active",
"New Mine (high intent)",
"Intermittent",
"Non-producing",
"Temporarily idled",
"Abandoned",
"Abandoned & sealed"
]
},
"default": [
"Active"
]
},
"mineTypes": {
"title": "Mine types",
"type": "array",
"description": "Filter by mine type. Facility = a mill / preparation plant (no extraction).",
"items": {
"type": "string",
"enum": [
"Surface",
"Underground",
"Facility"
],
"enumTitles": [
"Surface",
"Underground",
"Facility / plant"
]
}
},
"newMinesOnly": {
"title": "New mines only (high intent)",
"type": "boolean",
"description": "Only return operations with status New Mine — brand-new mining operations just registering with MSHA. The highest-intent leads (they're buying equipment, services, insurance and PPE right now).",
"default": false
},
"companyTypes": {
"title": "Company types",
"type": "array",
"description": "Filter by the operator's legal entity type.",
"items": {
"type": "string",
"enum": [
"Corporation",
"Limited Liability Corporation",
"Partnership",
"Sole Proprietor",
"Other"
],
"enumTitles": [
"Corporation",
"LLC",
"Partnership",
"Sole Proprietor",
"Other"
]
}
},
"minEmployees": {
"title": "Min employees",
"type": "integer",
"description": "Only operations reporting at least this many employees (a proxy for deal size). Leave blank for any."
},
"maxEmployees": {
"title": "Max employees",
"type": "integer",
"description": "Only operations reporting at most this many employees."
},
"operatorNameQuery": {
"title": "Operator / mine name contains",
"type": "string",
"description": "Only return operations whose operator name or mine name contains this text (case-insensitive)."
},
"controllerNameQuery": {
"title": "Parent controller contains",
"type": "string",
"description": "Only return operations whose parent / controlling company name contains this text — great for mapping every site owned by a target company."
},
"includeProduction": {
"title": "Include production & employment",
"type": "boolean",
"description": "Enrich each record with the latest reported year's average employment, coal tonnage (coal mines) and hours worked from MSHA's annual employment/production data. Cheap and recommended — gives a fresh activity & size signal.",
"default": true
},
"includeViolations": {
"title": "Include violations & penalties (compliance intel)",
"type": "boolean",
"description": "Enrich each emitted record with MSHA citation counts, significant-and-substantial (S&S) counts, total proposed penalties, amounts due and the last violation date. High-intent signal for safety, PPE, compliance, consulting and insurance vendors. Downloads MSHA's full national citation history (large), so these runs take longer — for the fastest results give the run 4096 MB or more memory (1+ vCPU). Records are charged at the compliance rate when this is on.",
"default": false
},
"violationsLookbackYears": {
"title": "Violations lookback (years)",
"minimum": 1,
"maximum": 20,
"type": "integer",
"description": "How many recent years count toward the 'recent violations' tally (the full all-time totals are always included).",
"default": 5
},
"maxResults": {
"title": "Max results",
"minimum": 1,
"type": "integer",
"description": "Maximum number of mine/operator records to return in this run.",
"default": 1000
},
"sortBy": {
"title": "Sort by",
"enum": [
"leadScore",
"employeesHigh",
"recentStatus",
"productionHigh",
"penaltiesHigh"
],
"type": "string",
"description": "Ordering of the returned records.",
"default": "leadScore"
},
"deduplicateResults": {
"title": "Deduplicate within a run",
"type": "boolean",
"description": "Skip duplicate mine IDs within this run (recommended).",
"default": true
},
"monitorMode": {
"title": "Monitoring mode — only new / changed mines",
"type": "boolean",
"description": "Remember which operations were already returned (in a named key-value store) and emit ONLY mines that are new or whose status changed since the last run — each tagged monitorEvent: \"new\" or \"status_change\". MSHA refreshes this data regularly; run on a Schedule to capture newly-registered operations as fresh leads. Works alongside Apify Schedules (the schedule starts the run; monitoring decides what's new).",
"default": false
},
"monitorKey": {
"title": "Monitor key",
"type": "string",
"description": "Names the memory used by monitoring mode. Use a distinct key per saved watch (e.g. per state/commodity) so different monitors don't share state.",
"default": "default"
},
"proxyConfiguration": {
"title": "Proxy configuration",
"type": "object",
"description": "Optional proxy. MSHA open data is downloaded directly and needs no proxy — leave this off for the fastest downloads. Enable Apify Proxy only if your network requires it.",
"default": {
"useApifyProxy": false
}
}
}
},
"runsResponseSchema": {
"type": "object",
"properties": {
"data": {
"type": "object",
"properties": {
"id": {
"type": "string"
},
"actId": {
"type": "string"
},
"userId": {
"type": "string"
},
"startedAt": {
"type": "string",
"format": "date-time",
"example": "2025-01-08T00:00:00.000Z"
},
"finishedAt": {
"type": "string",
"format": "date-time",
"example": "2025-01-08T00:00:00.000Z"
},
"status": {
"type": "string",
"example": "READY"
},
"meta": {
"type": "object",
"properties": {
"origin": {
"type": "string",
"example": "API"
},
"userAgent": {
"type": "string"
}
}
},
"stats": {
"type": "object",
"properties": {
"inputBodyLen": {
"type": "integer",
"example": 2000
},
"rebootCount": {
"type": "integer",
"example": 0
},
"restartCount": {
"type": "integer",
"example": 0
},
"resurrectCount": {
"type": "integer",
"example": 0
},
"computeUnits": {
"type": "integer",
"example": 0
}
}
},
"options": {
"type": "object",
"properties": {
"build": {
"type": "string",
"example": "latest"
},
"timeoutSecs": {
"type": "integer",
"example": 300
},
"memoryMbytes": {
"type": "integer",
"example": 1024
},
"diskMbytes": {
"type": "integer",
"example": 2048
}
}
},
"buildId": {
"type": "string"
},
"defaultKeyValueStoreId": {
"type": "string"
},
"defaultDatasetId": {
"type": "string"
},
"defaultRequestQueueId": {
"type": "string"
},
"buildNumber": {
"type": "string",
"example": "1.0.0"
},
"containerUrl": {
"type": "string"
},
"usage": {
"type": "object",
"properties": {
"ACTOR_COMPUTE_UNITS": {
"type": "integer",
"example": 0
},
"DATASET_READS": {
"type": "integer",
"example": 0
},
"DATASET_WRITES": {
"type": "integer",
"example": 0
},
"KEY_VALUE_STORE_READS": {
"type": "integer",
"example": 0
},
"KEY_VALUE_STORE_WRITES": {
"type": "integer",
"example": 1
},
"KEY_VALUE_STORE_LISTS": {
"type": "integer",
"example": 0
},
"REQUEST_QUEUE_READS": {
"type": "integer",
"example": 0
},
"REQUEST_QUEUE_WRITES": {
"type": "integer",
"example": 0
},
"DATA_TRANSFER_INTERNAL_GBYTES": {
"type": "integer",
"example": 0
},
"DATA_TRANSFER_EXTERNAL_GBYTES": {
"type": "integer",
"example": 0
},
"PROXY_RESIDENTIAL_TRANSFER_GBYTES": {
"type": "integer",
"example": 0
},
"PROXY_SERPS": {
"type": "integer",
"example": 0
}
}
},
"usageTotalUsd": {
"type": "number",
"example": 0.00005
},
"usageUsd": {
"type": "object",
"properties": {
"ACTOR_COMPUTE_UNITS": {
"type": "integer",
"example": 0
},
"DATASET_READS": {
"type": "integer",
"example": 0
},
"DATASET_WRITES": {
"type": "integer",
"example": 0
},
"KEY_VALUE_STORE_READS": {
"type": "integer",
"example": 0
},
"KEY_VALUE_STORE_WRITES": {
"type": "number",
"example": 0.00005
},
"KEY_VALUE_STORE_LISTS": {
"type": "integer",
"example": 0
},
"REQUEST_QUEUE_READS": {
"type": "integer",
"example": 0
},
"REQUEST_QUEUE_WRITES": {
"type": "integer",
"example": 0
},
"DATA_TRANSFER_INTERNAL_GBYTES": {
"type": "integer",
"example": 0
},
"DATA_TRANSFER_EXTERNAL_GBYTES": {
"type": "integer",
"example": 0
},
"PROXY_RESIDENTIAL_TRANSFER_GBYTES": {
"type": "integer",
"example": 0
},
"PROXY_SERPS": {
"type": "integer",
"example": 0
}
}
}
}
}
}
}
}
}
}