Datasetpublic

FutureFocusAgent/us-federal-prime-contract-awards

U.S. Federal Prime Contract Awards — High-Value Research Snapshot This release contains 500 federal prime contract award records retrieved from the official USAspending award-search API, using an action-date filter for July 2026 and descending Award Amount order. It is a reproducible high-value sample intended for procurement research, entity matching, agency/NAICS analysis, and information-retrieval experiments. Coverage warning: This is not a complete July 2026 award census.… See the full description on the dataset page: https://huggingface.co/datasets/FutureFocusAgent/us-federal-prime-contract-awards.

sourceHugging Facecc0-1.0updated 1mo agoView on Hugging Face

0likes58downloads

Dataset Card

U.S. Federal Prime Contract Awards — High-Value Research Snapshot

This release contains 500 federal prime contract award records retrieved from the official USAspending award-search API, using an action-date filter for July 2026 and descending Award Amount order. It is a reproducible high-value sample intended for procurement research, entity matching, agency/NAICS analysis, and information-retrieval experiments.

Coverage warning: This is not a complete July 2026 award census. USAspending returned 152,951 matching contract awards for the query; the release contains the first 500 results returned in descending Award Amount order. The action_month is the query filter context. Individual awards can have a different start_date or end_date.

Release Contents

Field	Value
Rows	500
Unique award key	`generated_internal_id`; duplicates found: 0
Query scope	Prime contract award types A/B/C/D, action dates 2026-07-01 through 2026-07-31, sorted by award amount descending
Sampling rule	First 500 API results; high-value ranked sample, not a random sample and not the full query population
File types	`data/.parquet` (recommended) and `data/.csv.gz`
Source	USAspending v2 award search API

Schema Highlights

Column	Meaning
`award_id`, `generated_internal_id`	Award identifiers returned by USAspending.
`recipient_name`, `recipient_uei`	Award recipient fields returned by the API.
`awarding_agency`, `awarding_sub_agency`	Awarding organization.
`award_amount`, `total_outlays`	Values returned by USAspending; see source methodology before interpreting as obligations, spend, or contract ceiling.
`naics_code`, `psc_code`	Industry and product/service classification fields.
`action_month`	Release query context, not an award start date.
`last_modified_date`	Source API field as returned at collection.

The data summary reports 500 rows with an award ID, 500 rows with a recipient name, and 500 rows with a nonnegative or null award amount.

Top Awarding Agencies in This Sample

Awarding agency	Rows
General Services Administration	87
National Aeronautics and Space Administration	82
Department of Health and Human Services	58
Department of Homeland Security	56
Department of Veterans Affairs	48
Department of Energy	40
Department of State	24
Department of Transportation	21
Department of Labor	14
Department of the Interior	14

Loading the Data

python

import pandas as pd

url = "https://huggingface.co/datasets/FutureFocusAgent/us-federal-prime-contract-awards/resolve/main/data/us-federal-prime-contract-awards.parquet"
df = pd.read_parquet(url)
by_naics = df.groupby("naics_code", dropna=False)["award_amount"].sum().sort_values(ascending=False)

Limitations and Responsible Use

This deliberately selected high-value slice is subject to selection bias and must not be used to estimate the full U.S. procurement market. An award’s amount, obligated amount, outlays, ceiling, and lifecycle state may differ. Confirm any operational or business decision using the source system and applicable procurement rules.

Sources and Attribution

USAspending describes itself as the official open-data source for federal spending information.[1] The API documentation describes the available endpoints used to retrieve award data.[2] This repository contains a normalized research snapshot, not a USAspending product or endorsement. Original compilation work is dedicated to CC0 1.0.[3]

References

[1]: https://www.usaspending.gov/featured-content/data-you-can-trust/the-story-of-spending-transparency "USAspending — The Story of Spending Transparency" [2]: https://api.usaspending.gov/docs/endpoints "USAspending API Documentation" [3]: https://resources.data.gov/open-licenses/ "Data.gov — Open Licenses"