{ "openapi": "3.0.1", "info": { "title": "USA Data.gov U.S. Government's Open Data Scrape", "description": "Stop wasting hours digging through thousands of government datasets. Our Data.gov scraper automatically gathers complete dataset details from the U.S. government's open data portal in minutes. Ideal for researchers, analysts, journalists, and teams needing reliable data without manual effort.", "version": "1.0", "x-build-id": "ABxaWivypJo7riT9x" }, "servers": [ { "url": "https://api.apify.com/v2" } ], "paths": { "/acts/parseforge~data-gov-scraper/run-sync-get-dataset-items": { "post": { "operationId": "run-sync-get-dataset-items-parseforge-data-gov-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": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK" } } } }, "/acts/parseforge~data-gov-scraper/runs": { "post": { "operationId": "runs-sync-parseforge-data-gov-scraper", "x-openai-isConsequential": false, "summary": "Executes an Actor and returns information about the initiated run in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$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": { "$ref": "#/components/schemas/runsResponseSchema" } } } } } } }, "/acts/parseforge~data-gov-scraper/run-sync": { "post": { "operationId": "run-sync-parseforge-data-gov-scraper", "x-openai-isConsequential": false, "summary": "Executes an Actor, waits for completion, and returns the OUTPUT from Key-value store in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$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": { "startUrl": { "title": "Start URL", "type": "string", "description": "Direct URL to scrape from Data.gov catalog. Use this OR search filters below, not both. Example: https://catalog.data.gov/dataset?q=climate" }, "searchQuery": { "title": "Search Query", "type": "string", "description": "Search term for datasets. Use this OR startUrl above, not both. Example: climate, healthcare" }, "topics": { "title": "Topics", "type": "array", "description": "Filter by topic groups. Select one or more topics from the dropdown. Leave empty for all topics.", "items": { "type": "string", "enum": [ "local", "climate5434", "older-adults-health-data", "energy9485" ], "enumTitles": [ "Local Government", "Climate", "Older Adults", "Energy" ] } }, "topicCategories": { "title": "Topic Categories", "type": "array", "description": "Filter by topic categories. Select one or more categories from the dropdown. Leave empty for all categories.", "items": { "type": "string", "enum": [ "Arctic", "Water", "Ecosystem Vulnerability", "Arctic Ocean, Sea Ice and Coasts", "Human Health", "Transportation", "Energy Infrastructure", "Food Resilience", "Atmospheric, Earth and Ocean Sciences", "Permafrost and Arctic Landscapes", "Arctic Weather and Extreme Events", "Melting Glaciers, Snow and Ice", "Coastal Flooding", "Arctic Peoples and Ecosystems", "Physical Data", "Food Production", "Food Safety and Nutrition", "Arctic Development", "Transportation Nodes", "Biodiversity", "Carbon Balance", "Energy Resources", "Food Security", "Arctic-Global", "Energy Supply", "Food Distribution", "Transportation Network", "Land Cover and Land Use", "Energy Conversion", "Ecosystem Services", "Hydrology", "Transportation Flow", "Water Resources", "Changing Oceans", "Climatology and Meteorology", "Asian", "Ecosystem Change Drivers", "Education", "Environment", "Hawaii", "Pacific Islands", "Winter Storm", "Arctic Oceans, Sea Ice and Coasts", "Atmospheric Chemistry", "Biology and Habitats", "Business", "Ecosystem" ], "enumTitles": [ "Arctic", "Water", "Ecosystem Vulnerability", "Arctic Ocean, Sea Ice and Coasts", "Human Health", "Transportation", "Energy Infrastructure", "Food Resilience", "Atmospheric, Earth and Ocean Sciences", "Permafrost and Arctic Landscapes", "Arctic Weather and Extreme Events", "Melting Glaciers, Snow and Ice", "Coastal Flooding", "Arctic Peoples and Ecosystems", "Physical Data", "Food Production", "Food Safety and Nutrition", "Arctic Development", "Transportation Nodes", "Biodiversity", "Carbon Balance", "Energy Resources", "Food Security", "Arctic-Global", "Energy Supply", "Food Distribution", "Transportation Network", "Land Cover and Land Use", "Energy Conversion", "Ecosystem Services", "Hydrology", "Transportation Flow", "Water Resources", "Changing Oceans", "Climatology and Meteorology", "Asian", "Ecosystem Change Drivers", "Education", "Environment", "Hawaii", "Pacific Islands", "Winter Storm", "Arctic Oceans, Sea Ice and Coasts", "Atmospheric Chemistry", "Biology and Habitats", "Business", "Ecosystem" ] } }, "datasetType": { "title": "Dataset Type", "enum": [ "", "geospatial" ], "type": "string", "description": "Filter by dataset type. Currently only 'geospatial' is available. Leave empty for all types." }, "tags": { "title": "Tags", "type": "array", "description": "Filter by tags. Enter one or more tag names (e.g., 'earth science', 'noaa', 'oceans'). Tags vary by dataset type, so use the text field for flexibility.", "items": { "type": "string" } }, "formats": { "title": "Formats", "type": "array", "description": "Filter by resource formats. Select one or more formats from the dropdown. Leave empty for all formats.", "items": { "type": "string", "enum": [ "XML", "ZIP", "HTML", "PDF", "CSV", "Esri REST", "TEXT", "JSON", "TIFF", "RDF", "SID", "BIN", "WMS", "EXCEL", "KML", "ArcGIS GeoServices REST API", "XYZ", "GeoJSON", "JPEG", "EXE", "PNG", "application/vnd.lotus-1-2-3", "WFS", "gml", "QGIS", "API", "WCS", "OGC WMS", "application/vnd.geo+json", "DOC", "GDB", "text/vnd.ascii-art", "application/geopackage+sqlite3", "NETCDF", "sos", "GPKG", "CDF", "SHP", "GZ", "Undefined", "nc", "TAR", "chemical/x-mdl-sdfile", "00", "application/html", "xsl", "cdfnc", "POWERPOINT" ], "enumTitles": [ "XML", "ZIP", "HTML", "PDF", "CSV", "Esri REST", "TEXT", "JSON", "TIFF", "RDF", "SID", "BIN", "WMS", "EXCEL", "KML", "ArcGIS GeoServices REST API", "XYZ", "GeoJSON", "JPEG", "EXE", "PNG", "application/vnd.lotus-1-2-3", "WFS", "gml", "QGIS", "API", "WCS", "OGC WMS", "application/vnd.geo+json", "DOC", "GDB", "text/vnd.ascii-art", "application/geopackage+sqlite3", "NETCDF", "sos", "GPKG", "CDF", "SHP", "GZ", "Undefined", "nc", "TAR", "chemical/x-mdl-sdfile", "00", "application/html", "xsl", "cdfnc", "POWERPOINT" ] } }, "organizationType": { "title": "Organization Type", "enum": [ "", "Federal Government", "State Government", "City Government", "University", "County Government" ], "type": "string", "description": "Filter by organization type. Select a type from the dropdown or leave empty for all types." }, "organization": { "title": "Organization", "enum": [ "", "allegheny-county-city-of-pittsburgh-western-pa-regional-data-center", "americorps-gov", "arlington-county", "usace-army-mil", "board-of-governors-of-the-federal-reserve-system", "city-and-county-of-durham-north-carolina", "city-of-austin", "city-of-baltimore", "city-of-baton-rouge", "city-of-bloomington", "city-of-boise", "city-of-charlotte", "city-of-chesapeake", "city-of-chicago", "city-of-ferndale-michigan", "city-of-hartford", "city-of-honolulu", "city-of-jackson-mississippi", "city-of-los-angeles", "city-of-new-orleans", "city-of-new-york", "city-of-philadelphia", "city-of-providence", "city-of-san-francisco", "city-of-santa-rosa", "city-of-seattle", "city-of-sioux-falls", "city-of-somerville", "city-of-tempe", "cdatribe-nsn-gov", "commodity-futures-trading-commission", "cfpb-gov", "cook-county-of-illinois", "usda-gov", "doc-gov", "dod-gov", "ed-gov", "doe-gov", "dhs-gov", "hud-gov", "doj-gov", "dol-gov", "state-gov", "doi-gov", "treasury-gov", "dot-gov", "va-gov", "district-of-columbia", "edac-unm-edu", "eac-gov", "eop-gov", "exim-gov", "fairfax-county-virginia", "fcc-gov", "fdic-gov", "fec-gov", "fgdc-gov", "fhfa-gov", "federal-laboratory-consortium", "federal-maritime-commission", "federal-retirement-thrift-investment-board", "federal-trade-commission", "florida-department-of-environmental-protection", "gsa-gov", "government-publishing-office", "gulf-coast-ecosystem-restoration-council", "imls-gov", "ibwc-gov", "king-county-washington", "lake-county-illinois", "legal-services-corporation", "library-of-congress", "loudoun-county-virginia", "louisville-metro-government", "mspb-gov", "mcc-gov", "montgomery-county-of-maryland", "municipality-of-anchorage", "nasa-gov", "nara-gov", "national-credit-union-administration", "nea-gov", "neh-gov", "nga-mil", "national-institute-of-standards-and-technology", "noaa-gov", "nsf-gov", "national-telecommunications-and-information-administration", "ntsb-gov", "nitrd-gov", "nrc-gov", "office-of-management-and-budget", "opm-gov", "opentopography", "pennsylvania-geospatial-data-clearinghouse", "pbgc-gov", "rrb-gov", "sandbox-organization", "cdip-ucsd-edu", "sec-gov", "sba-gov", "ssa-gov", "state-of-alaska", "arkansas-gov", "ca-gov", "state-of-connecticut", "state-of-hawaii", "state-of-illinois", "state-of-iowa", "state-of-maryland", "state-mn-us", "state-of-missouri", "state-of-new-york", "nc-gov", "state-of-oklahoma", "state-of-oregon", "state-of-washington", "federal-acquisition-service", "town-of-cary-north-carolina", "town-of-chapel-hill-north-carolina", "census-gov", "hhs-gov", "u-s-department-of-labor-bureau-of-labor-statistics", "epa-gov", "usgs-gov", "uidaho-edu", "u-washington-edu", "cpsc-gov", "eeoc-gov", "usitc-gov", "vcgi-org", "wake-county", "wvu-edu" ], "type": "string", "description": "Filter by organization. Select an organization from the dropdown or leave empty to search all organizations." }, "publisher": { "title": "Publisher", "enum": [ "", "ACF", "Agricultural Research Service", "Allegheny County", "Board of Governors of the Federal Reserve System", "Bureau of Justice Statistics", "Bureau of Labor Statistics", "Bureau of Transportation Statistics", "California Department of Education", "California Department of Public Health", "California Department of Technology", "California State Water Resources Control Board", "Centers for Disease Control and Prevention", "Centers for Medicare & Medicaid Services", "City of Philadelphia", "City of Seattle ArcGIS Online", "City of Tempe", "Department of Veterans Affairs", "District of Columbia Sentencing Commission", "Economic and Risk Analysis", "Economic Research Service, Department of Agriculture", "Employment and Training Administration", "Federal Highway Administration", "Federal Highway Administration (FHWA)", "Federal Housing Finance Agency", "Federal Motor Carrier Safety Administration", "FEMA/RESILIENCE/FIMA", "Food Safety and Inspection Service", "General Services Administration", "Health Resources and Services Administration", "HIFLD", "ICE", "Institute of Museum and Library Services", "Justice Management Division", "Lake County Illinois GIS", "Metropolitan Police Department", "Mine Safety and Health Administration", "National Aeronautics and Space Administration", "National Agricultural Statistics Service, Department of Agriculture", "National Center for Education Statistics (NCES)", "National Highway Traffic Safety Administration", "National Institute of Standards and Technology", "National Institutes of Health (NIH)", "National Library of Medicine", "National Renewable Energy Laboratory", "National Science Foundation", "NCO NITRD", "Office for Civil Rights (OCR)", "Office of Analysis and Research Services", "Office of Debt Management", "Office of Federal Student Aid (FSA)", "Office of Tariff Affairs and Trade Agreements", "Office of the Chief Economist (OCE)", "Office of the Chief Technology Officer", "Office of the Secretary of Transportation", "Public Affairs", "Rural Development, Department of Agriculture", "Small Business Administration", "Social Security Administration", "State of New York", "Structured Disclosure", "Substance Abuse and Mental Health Services Administration", "U.S. Census Bureau", "U.S. Commodity Futures Trading Commission", "U.S. Department of Housing and Urban Development", "U.S. Energy Information Administration", "U.S. EPA Office of Environmental Information (OEI)", "U.S. EPA Office of Research and Development (ORD)", "U.S. Food and Drug Administration", "U.S. Geological Survey", "United States Mint, Financial Department (CFO)", "University of Maryland (UMD)", "Unspecified" ], "type": "string", "description": "Filter by publisher. Select a publisher from the dropdown or leave empty to search all publishers." }, "bureau": { "title": "Bureau", "enum": [ "", "009:70", "009:33", "005:45", "005:18", "007:57", "015:13", "029:25", "011:14", "015:20", "010:76", "012:20", "010:04", "010:06", "010:22", "006:07", "009:20", "009:38", "202:00", "007:97", "007:00", "018:00", "019:00", "009:00", "025:00", "011:00", "014:00", "015:00", "021:00", "029:00", "019:60", "029:40", "012:25", "006:05", "010:84", "015:05", "011:12", "006:06", "005:13", "006:08", "012:05", "019:20", "019:10", "020:00", "350:00", "005:49", "021:12", "011:10", "356:00", "015:11", "021:15", "021:17", "011:20", "021:27", "021:36", "015:04", "015:12", "009:10", "005:84", "005:96", "009:15", "018:50", "015:60", "015:45", "006:25", "011:05", "007:05", "026:00", "021:18", "006:55", "009:25", "019:05", "010:24", "422:00", "011:08", "006:60", "005:53", "007:17", "018:10", "018:15", "018:45", "018:12", "011:21", "018:40", "018:20", "021:04", "007:10", "200:00", "021:50", "007:15", "011:06", "005:55", "001:05", "009:30", "006:51", "010:18", "010:12", "015:25", "029:15" ], "type": "string", "description": "Filter by bureau. Select a bureau from the dropdown or leave empty to search all bureaus." }, "location": { "title": "Location", "type": "string", "description": "Filter by location name. Example: New York, California. When using startUrl, location parameters (ext_location, ext_bbox, ext_prev_extent) are preserved automatically." }, "sort": { "title": "Sort", "enum": [ "views_recent desc", "score desc, metadata_modified desc", "title_string asc", "title_string desc", "metadata_modified desc", "metadata_created desc" ], "type": "string", "description": "Sort order. Default: Popular (views_recent desc). Options: views_recent desc (Popular), score desc, metadata_modified desc (Relevance), title_string asc (Name A-Z), title_string desc (Name Z-A), metadata_modified desc (Last Modified), metadata_created desc (Date Added)", "default": "views_recent desc" }, "maxItems": { "title": "Max Items", "minimum": 1, "maximum": 1000000, "type": "integer", "description": "Free users: Limited to 100. Paid users: Optional, max 1,000,000" } } }, "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 } } } } } } } } } }