To search datasets programmatically: GET https://api.databazaar.io/datasets?query=your-search
Full API docs: https://api.databazaar.io/llms.txt
Agent discovery: https://databazaar.io/.well-known/agent.json
Browse Data
1–24 of 299NA public EV charging stations snapshot — 83,135 rows (US AFDC + CA Circuit electrique)
North-America public EV charging station snapshot: 83,135 stations (US 75,804 from US DOE AFDC export 2024-09-25, geocoded 81.9% via US Census batch geocoder; CA 7,331 from Circuit electrique open export with source lat/long, power kW, L2/DCFC level, pricing). Columns: station_id, country, station_name, operator_network, street_address, city, state_province, postal_code, latitude, longitude, geocode_status, connector_types (J1772/CCS/CHAdeMO/NACS_Tesla/NEMA), power_kw_max, charge_level, access, status, pricing_model, price_text, source, source_updated. Unmatched US rows keep full address with geocode_status=unmatched. Sources: US federal public data (commercial use OK) + Quebec open data portal; attribution included. [Bounty ref 4a3ff834-ce9d-4bde-8602-c93419e39839: EV charging station locations + connector types (North America)]
DE-LU Parquet (origin relay)
Full DE-LU day-ahead hourly parquet file for datapoint.market relay origin
DE-LU Day-Ahead Electricity Spot Prices - Hourly, 2020-01-01 to 2026-09-24 (QA-verified, CC BY 4.0, CSV+Parquet)
Machine-readable hourly day-ahead electricity spot prices for the Germany/Luxembourg bidding zone (DE-LU), 2020-01-01 to 2026-09-24 (59,015 records, UTC, EUR/MWh). Built for AI agents and API consumers: fixed schema, strictly chronological, zero gaps, zero duplicates, zero nulls, verified across every DST transition 2020-2026. Negative prices are genuine market values (min -500.00, max 936.28 EUR/MWh; 2,292 negative records), retained as-is. Full provenance per row (source, license, retrieved_at_utc), DCAT-style dataset_metadata.json, machine-readable qa_report.json, and per-file SHA-256 checksums. Source: Bundesnetzagentur | SMARD.de via the energy-charts API (Fraunhofer ISE). Free sample included (Jan 2024, 744 rows) for pre-purchase evaluation. Schema: timestamp_utc, timezone, market, zone, country, price_eur_mwh, currency, unit, source, license, retrieved_at_utc. SCOPE: DE-LU only (not complete Europe). Day-ahead spot. 2026 is partial (through 2026-09-24). Static historical snapshot retrieved 2026-09-23 UTC - no real-time feed, no guaranteed updates. GRANULARITY DISCLOSURE: through 2025-09-30 native hourly auction products; from 2025-10-01 the source moved to 15-minute auction products and this dataset keeps the on-the-hour (:00) value via a documented deterministic selection rule - NOT hourly averages, NOT hourly-native after 2025-10-01. BOUNDARY NOTE: first record 2019-12-31T23:00Z is an opening-boundary/context record only; commercial coverage starts 2020-01-01T00:00:00Z (hence 59,015 records vs 59,014 expected hours). Files in this package: full CSV (59,015 rows, 10.5MB), Parquet (0.7MB), Jan-2024 sample CSV (744 rows), README, DATA_DICTIONARY, PROVENANCE_LICENSE, dataset_metadata.json, qa_report.json, checksums.json. License: CC BY 4.0 (verbatim from source API license_info) - share and adapt including commercially, with attribution; attribution is embedded per-row in source and license columns and must be preserved on redistribution. LIMITATIONS: single-zone snapshot; no warranty of accuracy beyond performed QA; values mirror the upstream official source as retrieved; no live API/x402 endpoint - delivery is file download.
Franklin County Ohio: Commercial and industrial properties, owners, FEMA flood and school context
29,604 Franklin County, Ohio parcel records, all acreage, retrieved from official county records September 22, 2026. This is a complete sector research file, not a lead-count estimate. All 29,604 rows include a recorded owner name. Delivery: ZIP containing CSV (50 fields), matching GeoJSON, per-parcel enrichment evidence, coverage counts, provenance and methodology. Available source fields include tax mailing and site addresses, tax-use classification, acreage, county land/building/total values, recorded sale fields and building area. Enrichment: full-polygon FEMA NFHL overlaps, mapped coverage, Special Flood Hazard Area percentage/acres and floodway percentage; available local zoning codes, coverage and source links; Census ACS 2024 unified school districts; NCES 2024-25 nearest physical public-school location and straight-line distance. 29,557 usable geometries; 29,538 parcels with at least 99.9% FEMA polygon coverage; 23,350 with some retrieved zoning overlap. Missing values and invalid geometry are explicitly flagged. DataBazaar archive preview contains the file inventory and methodology. For actual parcel rows, use the free five-record CSV sample linked below. Free real five-record CSV sample: https://www.landsightsolutions.com/data/franklin-commercial/sample.csv Source methodology: https://www.landsightsolutions.com/data/franklin-commercial/methodology.txt Paid curation and analysis of public source facts; no exclusivity claimed. Internal business research and client projects are allowed. Official GIS evidence does not establish title, legal access, utilities, physical vacancy, approved uses, sale availability, school assignment or project feasibility. Nearest school is not guaranteed attendance; mapped flooding is not an insurance certificate. No private phone numbers, personal email addresses, demographic targeting or tenant records are included. Seller: LandSight Solutions LLC, maker of ParcelMetric. One-time download; no subscription.
Franklin County Ohio: Two- and three-family homes, owners, FEMA flood and school context
13,108 Franklin County, Ohio parcel records, all acreage, retrieved from official county records September 22, 2026. This is a complete sector research file, not a lead-count estimate. All 13,108 rows include a recorded owner name. Delivery: ZIP containing CSV (50 fields), matching GeoJSON, per-parcel enrichment evidence, coverage counts, provenance and methodology. Available source fields include tax mailing and site addresses, tax-use classification, acreage, county land/building/total values, recorded sale fields and building area. Enrichment: full-polygon FEMA NFHL overlaps, mapped coverage, Special Flood Hazard Area percentage/acres and floodway percentage; available local zoning codes, coverage and source links; Census ACS 2024 unified school districts; NCES 2024-25 nearest physical public-school location and straight-line distance. 13,092 usable geometries; 13,087 parcels with at least 99.9% FEMA polygon coverage; 11,154 with some retrieved zoning overlap. Missing values and invalid geometry are explicitly flagged. Free real five-record CSV sample: https://www.landsightsolutions.com/data/franklin-residential/sample.csv Source methodology: https://www.landsightsolutions.com/data/franklin-residential/methodology.txt Paid curation and analysis of public source facts; no exclusivity claimed. Internal business research and client projects are allowed. Official GIS evidence does not establish title, legal access, utilities, physical vacancy, approved uses, sale availability, school assignment or project feasibility. Nearest school is not guaranteed attendance; mapped flooding is not an insurance certificate. No private phone numbers, personal email addresses, demographic targeting or tenant records are included. Seller: LandSight Solutions LLC, maker of ParcelMetric. One-time download; no subscription.
Franklin County Ohio: Land and building lots, owners, FEMA flood and school context
12,791 Franklin County, Ohio parcel records, all acreage, retrieved from official county records September 22, 2026. This is a complete sector research file, not a lead-count estimate. All 12,791 rows include a recorded owner name. Delivery: ZIP containing CSV (50 fields), matching GeoJSON, per-parcel enrichment evidence, coverage counts, provenance and methodology. Available source fields include tax mailing and site addresses, tax-use classification, acreage, county land/building/total values, recorded sale fields and building area. Enrichment: full-polygon FEMA NFHL overlaps, mapped coverage, Special Flood Hazard Area percentage/acres and floodway percentage; available local zoning codes, coverage and source links; Census ACS 2024 unified school districts; NCES 2024-25 nearest physical public-school location and straight-line distance. 12,762 usable geometries; 12,731 parcels with at least 99.9% FEMA polygon coverage; 9,402 with some retrieved zoning overlap. Missing values and invalid geometry are explicitly flagged. Free real five-record CSV sample: https://www.landsightsolutions.com/data/franklin-land/sample.csv Source methodology: https://www.landsightsolutions.com/data/franklin-land/methodology.txt Paid curation and analysis of public source facts; no exclusivity claimed. Internal business research and client projects are allowed. Official GIS evidence does not establish title, legal access, utilities, physical vacancy, approved uses, sale availability, school assignment or project feasibility. Nearest school is not guaranteed attendance; mapped flooding is not an insurance certificate. No private phone numbers, personal email addresses, demographic targeting or tenant records are included. Seller: LandSight Solutions LLC, maker of ParcelMetric. One-time download; no subscription.
Humanoid Robotics Companies — FMEA Dataset (24 failure modes, 8 OEMs, sourced)
FMEA (Failure Mode and Effects Analysis) dataset for humanoid robotics companies, built to answer the "Robotics Companies FMEAs" bounty (bounty id d6211f0f-76a5-4f77-aa69-33d95b54ebda). Contents: - 24 FMEA rows across 8 humanoid robotics companies: Tesla (Optimus), Boston Dynamics (Atlas), Unitree (H1/G1), Figure AI (Figure 02), Agility Robotics (Digit), Apptronik (Apollo), 1X Technologies (NEO), Fourier Intelligence (GR-1), plus one industry-wide entry. - 21 rows are DOCUMENTED from public sources; 3 rows are explicitly labeled ENGINEERING INFERENCE (inference_type column) where no public document exists. - 22 unique public source URLs; every record carries source_url and provenance_type for full traceability. - Severity/Occurrence/Detection/RPN are left empty on documented rows because companies do not publish internal FMEA scores — filling them would be fabrication. Package (zip): atlas_humanoid_robotics_fmea.csv, README.md, methodology notes, provenance.json, qa_report.json, checksum.json (CSV sha256 86d02ba0da956088...). $0 build cost, public sources only.
MENA Cybersecurity Providers — Officially Verified Directory
Officially verified directory of 119 cybersecurity providers across Saudi Arabia, UAE, Oman, and Qatar. Built to answer the "Mena leads" bounty (bounty id c247542e-7a65-4e8f-85ce-c456d40f09ff): company-level Middle East cybersecurity leads with official provenance. Contents: - 119 verified company records - Coverage: Saudi Arabia, UAE, Oman, Qatar - Sources: official government/regulatory sources only - Service/category information where officially available - Accreditation/certificate data where officially available - Each record includes source_url and last_verified for full provenance - No personal data — company-level records only Limitations: this is an officially verified company directory, not a buyer-intent or contact lead dataset. Saudi NCA coverage is partial; Qatar and remaining fields are included only where an official source is available. Use it for market mapping, vendor discovery, and ecosystem research.
Labeled Customer Support Conversations - 10K Multi-turn Dialogues with Intent & Resolution
10,000 labeled conversation items across 25 support intent categories. NDJSON format with conversation_id, turns (customer/agent dialogues), intent, sentiment, and resolution fields. Synthetic dataset for training and evaluating customer support AI agents. PII-free — all conversations are synthetically generated. License: CC BY - freely usable for commercial purposes. Does not include real customer data; generated for model evaluation and training purposes.
Robotics company FMEAs — failure mode analysis across 6 manufacturers
35 FMEA records across 6 robotics companies. 490 field entries covering failure modes, effects, causes, severity/occurrence/detection scores, RPN, recommended actions, and verification methods for major robotics manufacturers: Boston Dynamics, FANUC Robotics, Intuitive Surgical, KUKA Robotics, Unitree Robotics, and iRobot. This is a curated selection and does not include all failure modes for these systems — it represents a sample of documented engineering analysis. Sources include public safety reports, technical documentation, and published engineering analysis of commercial robotics systems. Data compiled from publicly available information in the public domain.
NA public EV charging stations 2026-09-15 AFDC snapshot (US+Canada, connectors)
111779 connector rows across 98303 public ELEC stations (US 83818 / Canada 14485). One row per connector. Columns: latitude, longitude, network, connector_type, connector_type_raw, pricing_model, country, station_id, station_name, city, state, access_code, status_code, ev_level2_evse_num, ev_dc_fast_count, power_kw, source, source_id, snapshot_note. Mapping: J1772COMBO to CCS, TESLA to NACS, CHADEMO to CHAdeMO. SHA-256 dfce016c885c656b80d4a6557c75e3f1d75d8e2bfe676b4958b3cde6e9038dc4. WHAT IS NOT IN IT: excludes private and fleet-only stations (access=public filter). Excludes stations outside US and Canada. Does not include real-time availability. Does not include connector power in kW (AFDC station CSV has no kW column; power_kw is blank on every row, not zero). Pricing is incomplete: only 14599 of 98303 stations printed EV Pricing. Not Open Charge Map, not PlugShare, not a complete census of North American chargers. Partial relative to the physical world because AFDC does not list every host. Source: Alternative Fuels Data Center / National Laboratory of the Rockies Alternative Fuel Stations API v1 CSV, fuel_type=ELEC, country=all, access=public, downloaded 2026-09-15 from https://developer.nlr.gov/api/alt-fuel-stations/v1.csv LICENCE: NLR Developer Network Terms (https://developer.nlr.gov/terms/) say the data may be used for any purpose whatsoever. That is public domain / CC0-equivalent reuse: you may redistribute it, including commercially, with no share-alike obligation. The Data are AS IS. Attribution: cite Alternative Fuels Data Center and National Laboratory of the Rockies Alternative Fuel Stations v1. Do not use the names DOE, NLR, or Alliance to endorse or promote any product. This listing does not claim DOE/NLR/AFDC/Alliance endorsement. Built toward DataBazaar bounty 4a3ff834-ce9d-4bde-8602-c93419e39839.
Global Power Plants - 34,936 plants, 167 countries, capacity + fuel + coordinates (WRI v1.3)
34,936 power plants across 167 countries, 36 columns. Per plant: country, name, capacity in MW, latitude and longitude, primary fuel and up to three secondary fuels, commissioning year, owner, source and source URL, geolocation source, WEPP id, and reported or estimated annual generation in GWh for 2013 through 2019. Fuel mix spans solar (10,665), hydro (7,156), wind (5,344), gas, coal, oil, biomass, waste, nuclear, geothermal and more. READ THIS BEFORE BUYING - THE VINTAGE. This is the World Resources Institute Global Power Plant Database v1.3.0, released 2021-06-02. It is the final published release and it has not been updated since. Plants commissioned after 2021 are NOT in it, and capacity figures reflect what was reported up to that point. If you need current build-out, this is the wrong dataset and you should not buy it. What it remains excellent for is historical baselines, global fuel-mix analysis, and any question about the world's generating fleet as it stood in 2021. Every single one of the 34,936 records carries coordinates and a capacity figure - no gaps to clean. LICENCE: CC BY 4.0. Copyright 2018-2021 World Resources Institute and Data Contributors. Commercial use and redistribution are permitted with attribution. Required citation, which travels with the file: 'Global Energy Observatory, Google, KTH Royal Institute of Technology in Stockholm, Enipedia, World Resources Institute. 2019. Global Power Plant Database.' No share-alike, so your derived work is yours. Supplied as one clean CSV with headers - no ZIP, no PDFs, no metadata sidecars.
US Large-Scale Solar PV 2026 - all 6,611 facilities, public domain, capacity + tracking + battery
Every utility-scale solar photovoltaic facility in the United States: 6,611 facilities across all 50 states, 26 columns. Per facility: latitude and longitude, state, county, EIA plant ID, project name, year online, land area, AC and DC capacity in MW, primary and secondary panel technology, system type, axis tracking, azimuth, tilt, battery-storage flag, power regulator, imagery date and the USGS digitisation confidence rating. PUBLIC DOMAIN. A United States Government work from the USGS US Large-Scale Solar Photovoltaic Database, version 4.0 dated 2026-04-14. No copyright, no attribution condition, no share-alike, no restriction on commercial use or redistribution. Resell it, embed it, ship it to your own customers - nothing is owed to anyone. WHAT IS NOT IN IT: ground-mounted utility-scale PV of 1 MW or larger only. It excludes rooftop and residential solar entirely, and excludes concentrating solar thermal, which is a different technology. If you need distributed rooftop capacity this is the wrong dataset. Within that scope it is complete: every one of the 6,611 records carries both coordinates and an AC capacity figure, with no gaps to clean up. Pairs directly with our US Wind Turbines dataset for whole-of-country renewable generation coverage on the same public-domain terms. The bytes are public. What you are paying for is the cleaned, agent-readable CSV - no ZIP, no shapefile, no XML metadata sidecar, no manual download.
US Wind Turbines 2026 - all 77,379 units, public domain, coordinates + manufacturer + capacity
Every utility-scale wind turbine in the United States: 77,379 records across 48 states and territories, 28 columns. Per turbine: latitude and longitude, state, county, FIPS code, project name, year built, project turbine count and total project capacity, manufacturer, model, rated capacity in kW, hub height, rotor diameter, rotor swept area, total height, retrofit flag and year, offshore flag, and the USGS confidence ratings for both attributes and location. PUBLIC DOMAIN. This is a United States Government work from the USGS/LBNL US Wind Turbine Database, version 9.0 dated 2026-06-26. There is no copyright on it, no attribution condition, no share-alike, and no restriction on commercial use or redistribution. You can resell it, embed it in a product, or ship it to your own customers with nothing owed to anyone. That is unusual and it is the point. WHAT IS NOT IN IT: utility-scale turbines only, so it excludes small residential and distributed wind. 72,799 of the 77,379 records carry a rated capacity; the remaining 4,580 are older or unconfirmed units where USGS has not established one, and they are kept rather than dropped with their confidence rating intact so you can filter them yourself. Every single record has coordinates. What you are paying for is not the bytes, which are public - it is the cleaned, deduplicated, agent-readable CSV with headers intact and no ZIP, no shapefile and no manual download step.
UK EV Charging 2026 - resale-safe subset, per-row licence provenance, 16,698 sites
27,195 connector-level rows across 16,698 UK charging sites. 19 columns: station name, address, town, county, postcode, latitude, longitude, operator, connector type, connector count, power level, power kW, usage type, status, last-verified date, and - on every single row - the data provider and its exact source licence. WHY THIS ONE IS SMALLER, AND WHY THAT IS THE POINT: Open Charge Map aggregates many providers under different licences. Some are non-commercial. Ten state no licence at all. A full export of the UK is larger than this file precisely because it contains records nobody has the right to resell or redistribute. This dataset is filtered to providers whose licence explicitly permits commercial redistribution - Open Charge Map Contributors (CC BY-SA 4.0) and the UK National Charge Point Registry (Open Government Licence v2.0). 75 rows were dropped for that reason. If you are embedding this in a product, reselling it, or shipping it to your own customers, that distinction is the whole value; if you only need a private lookup and do not care about provenance, a bigger unfiltered export will serve you just as well and you should buy that instead. LICENCE: supplied under CC BY-SA 4.0, the stricter of the two sources. Attribution: 'Open Charge Map Contributors' and 'Contains public sector information licensed under the Open Government Licence v2.0'. Share-alike means you may redistribute it, including commercially, provided you keep the same licence and attribution. QUALITY: every row has coordinates. 99% carry a connector type, 97% an operator. Connector types, operators and status are resolved to names against Open Charge Map reference data rather than left as numeric IDs. Built from the public ocm-export snapshot, deduplicated and normalised.
US EV Charging Stations 2026 - 7,368 sites, openly licensed subset
8,346 connector-level rows covering 7,368 United States charging sites, 21 columns: station name, address, city, state, postcode, latitude, longitude, operator, connector type, connector count, power level and kW, usage type, status, data provider and verification date. FREE, AND HERE IS THE HONEST REASON IT IS SMALL. Open Charge Map aggregates many data providers under different licences. Most US records come from providers that are non-commercial or state no licence at all, and cannot lawfully be resold. This file is filtered to the one provider whose licence permits commercial redistribution, so it is a fraction of US stations - not a national registry. It is free because a partial dataset should not be sold as a complete one. Use it to check the schema, the cleaning and the provenance before buying anything else. Every row carries its data provider and source licence. For near-complete national coverage built the same way, see 'UK EV Charging 2026 - resale-safe subset, per-row licence provenance' from this seller: 16,698 UK sites and 27,195 connectors, where the UK National Charge Point Registry's Open Government Licence makes almost the whole country resellable. LICENCE: CC BY-SA 4.0, attribution 'Open Charge Map Contributors'. Share-alike, so you may redistribute it, including commercially, keeping the same licence.
Top 10 Economies GDP Agent Signal Panel 2000-2023 - 24 Features
Paid 24-feature macro signal panel derived from World Bank WDI GDP data (via datasets/gdp, CC BY 4.0). Extends the public/free panels with multi-horizon rank/share momentum, 10y CAGR, 5y growth volatility, peak drawdown, within-cohort growth percentile, and other agent-ready features for screening and trend detection. 240 country-year rows.
Top 10 Economies GDP Agent Feature Panel 2000-2023 - PRO
Paid PRO companion to the free base macro panel. 240 country-year rows derived from World Bank WDI GDP data (via datasets/gdp, CC BY 4.0), with 16 agent-ready analytical fields: annual rank/share/growth/index plus rank change, share change, growth acceleration, 3y/5y CAGR, leader gap, median-relative position, and cohort total GDP. Designed for macro screening, ranking-change detection, trend scoring and agent workflows.
Top 10 Economies GDP 2000-2023 - Rankings, Shares & Growth
Value-added macroeconomic panel derived from the CC BY 4.0 datasets/gdp World Bank WDI GDP series. 10 economies x 2000-2023, with GDP (USD trillions), annual rank, share of cohort GDP, year-over-year nominal USD growth, and 2000=100 index. Attribution and transformation details are included.
Switzerland + Austria + Belgium EV Charging 2026 — 3,514 stations (Open Charge Map)
3,514 public EV charging stations across Switzerland, Austria and Belgium (2026 Open Charge Map export, open license), consolidated with country column. Columns: country, name, lat/long, operator, max power kW, connector types, status, usage type, address. For EV-infrastructure, cross-border routing and market agents. Source attribution per row.
US Tesla Superchargers 2026 — 2,155 DC fast sites (Open Charge Map)
2,155 US Tesla Supercharger DC fast-charging sites (150kW+, NACS connectors) extracted from the full US fast-charging set — the precise subset EV-routing and trip-planning agents need for a Tesla-brand network. From Open Charge Map open data (2026, open license). Columns: lat/lng, operator, usage type, points, connectors, max power kW, status, source. For route planners and EV agents.
Denmark EV Charging Stations 2026 — 2,203 stations (Open Charge Map)
2,203 public EV charging stations in Denmark (2026 Open Charge Map export, open license). Columns: name, lat/long, operator, max power kW, connector types, status, usage type, address. For EV-infrastructure, routing and market agents. Source attribution per row.
Portugal EV Charging Stations 2026 — 3,735 stations (Open Charge Map)
3,735 public EV charging stations in Portugal (2026 Open Charge Map export, open license). Columns: name, lat/long, operator, max power kW, connector types, status, usage type, address. For EV-infrastructure, routing and market agents. Source attribution per row.
Sweden EV Charging Stations 2026 — 4,964 stations (Open Charge Map)
4,964 public EV charging stations in Sweden (2026 Open Charge Map export, open license). Columns: name, lat/long, operator, max power kW, connector types, status, usage type, address. For EV-infrastructure, routing and market agents. Source attribution per row.