The dataset is currently empty. Upload or create new data files. Then, you will be able to explore them in the Dataset Viewer.
Top federal award recipients (USAspending — 10k entities by award $$$)
Published by MORVS — source-linked public data for AI agents. Catalog: morvs.ai/regulatory · Dataset page: api.morvs.ai/datasets/fed-recipients
Top 10,000 federal award recipients ranked by total dollars received. Includes contracts, grants, loans across DOD, HHS, NSF, NIH, USDA, and all civilian agencies. Examples: Lockheed Martin $49.6B (defense), CA Dept of Healthcare Services $112.3B (Medicaid), NY State Dept of Health $75.6B, plus universities (Johns Hopkins, MIT, Stanford), hospitals, nonprofits, and corporations. Each entity has the USAspending Recipient ID, UEI (modern Unique Entity Identifier replacing DUNS), entity type (corporation / government / university / nonprofit / healthcare), total lifetime award dollars, and rank. schema.org/Organization JSON-LD with UEI + DUNS identifiers. Source: api.usaspending.gov/api/v2/recipient/ (public POST API, no auth).
Live API
This dataset is served via a live REST API at api.morvs.ai. The card you're reading exists so HuggingFace's index can route AI agents + researchers to the canonical source.
- API endpoint:
https://api.morvs.ai/api/v1/fed-recipients/top - Dataset landing page: https://api.morvs.ai/datasets/fed-recipients
- DataCatalog (JSON-LD): https://api.morvs.ai/.well-known/dataset.json
- DCAT-US (data.gov-compatible): https://api.morvs.ai/data.json
Quick start
# Recent records (JSON)
curl 'https://api.morvs.ai/api/v1/fed-recipients/top?limit=10'
# Browse on the web
open 'https://api.morvs.ai/datasets/fed-recipients'
MCP (Model Context Protocol)
This dataset is also accessible via MCP at https://api.morvs.ai/mcp — Claude, ChatGPT, and other MCP-compatible agents can call it directly.
Source
- Agency: USAspending.gov (Treasury Bureau of the Fiscal Service)
- Primary source: https://www.usaspending.gov/recipient
- License: CC0 1.0 applies only to MORVS-authored metadata in this card. Underlying records authored by US federal employees as part of their official duties are US federal government works not subject to copyright under 17 U.S.C. § 105. See MORVS source terms for third-party and other material. Free or keyless access does not alter source terms.
- Update cadence: Varies by source; see the live freshness report
Keywords
federal contractor, federal grant recipient, USAspending, Lockheed Martin, Boeing, Raytheon, UEI, DUNS, federal contract, DOD contractor
Citation
@misc{ai_analytics_fed_recipients_2026,
title = {{name}},
author = {MORVS},
year = {2026},
url = {https://api.morvs.ai/datasets/fed-recipients},
note = {Derived from USAspending.gov (Treasury Bureau of the Fiscal Service). CC0 applies only to MORVS-authored card metadata; federal works follow 17 U.S.C. 105.}
}
Related
- Full data catalog
- Unified compliance screening (30+ enforcement lists)
- Cross-vertical entity timeline
- OpenAPI 3.1 spec
- llms.txt for AI agents
Maintained by MORVS (operated by AI Analytics LLC). CC0 applies only to MORVS-authored metadata in this card; please cite the upstream agency.
- Downloads last month
- 27