buriedsignals/anomaly/usaspending-awards
Claude Code and agent skill by buriedsignals/anomaly
- community
Checking that the SKILL.md file is still in the repository.
Use this skill to search USAspending prime federal contracts, grants, loans, IDVs, direct payments, or other assistance by recipient, keyword, award ID, and time window. Apply it to award-level procurement and assistance research; keep award amount distinct from transactions, obligations, outlays, payments, and subawards, and resolve recipients with UEI and award identifiers.
Install
npx vskill@latest install buriedsignals/anomaly/usaspending-awardsSKILL.md
Query USAspending federal awards
meta.yaml is the executable contract. Use only released operations.
<!-- BEGIN GENERATED OPERATION STATUS -->
Current catalogue release status
meta.yaml is authoritative. Execute only operations listed as Released below; other sections in this playbook may document provider or unreleased adapter scope.
Released
us/usaspending/awards:search-awards— Search one award-type group over an explicit date range or federal fiscal year with bounded, type-compatible fields and pagination.
<!-- END GENERATED OPERATION STATUS -->
Define type and time explicitly
A federal fiscal year runs October 1 through September 30. Supplying explicit dates requires both start_date and end_date. If neither is present, the adapter uses the current federal fiscal year and returns that window in query_window.
Use recipient to search recipient name, UEI, and DUNS. Use keyword for the broader award keyword filter. award_id is quoted upstream for exact matching.
Interpret amounts and records
amountis Award Amount, except loans where it is Loan Value.total_outlaysandsubsidy_costare separate fields when valid for the type.- This operation returns prime awards, not transactions or subawards.
- A recipient-name match is not entity resolution; verify UEI, generated award
ID, agency, dates, and the linked award profile.
- Corrections and modifications can change award-level values over time.
Fill the procurement contract
procurement-processes and procurement-awards run the contract search, then read each award's detail (/api/v2/awards/{generated_unique_award_id}/, one extra request per record) and return rows of the procurement domain contract (domains/procurement.yaml). Write each to CSV and run the procurement detectors with python -m anomaly.oneshot --tables procurement_processes=... --tables procurement_awards=....
| If the source publishes | map it to column | |
|---|---|---|
generated_unique_award_id | process_id and award_id (one award per contract action) | |
awarding_agency.subtier_agency.code / .name | buyer_id, buyer_name (office_agency_name is finer if needed) | |
| (federal data) | country = US, currency = USD | |
latest_transaction_contract_data.naics (PSC in product_or_service_code, kept as psc) | category_code | |
latest_transaction_contract_data.extent_competed: A → open; D, CDO → selective; F → limited; B, C, E, G, NDO → direct | procurement_method (raw code kept as extent_competed) | |
latest_transaction_contract_data.number_of_offers_received | number_of_tenderers | |
recipient.recipient_uei, recipient.recipient_name | supplier_id, supplier_name | |
total_obligation | award_value | |
date_signed | award_date | |
base_exercised_options (current value), base_and_all_options (potential value) | extra columns, not mapped | |
| solicitation dates (SAM.gov, not USAspending) | tender_published_date, tender_deadline stay empty |
The extent-competed mapping follows the FPDS code definitions in the USAspending data dictionary and the OCDS method codelist; F (simplified acquisition, competed) is limited because the buyer chooses whom to ask.
final_value stays empty: the API reports only current values, not the value at award, and cumulative obligations grow with incremental funding of an unchanged contract (a live sample on 2026-10-27 flagged 47 of 50 contracts that way). For value_growth_after_award, take the bulk award download (Contracts_PrimeTransactions) and map the first action's base_and_exercised_options_value to award_value and the latest base_and_exercised_options_value to final_value.
Read the API reference for filters, type codes, type-specific fields, pagination, and amount semantics. Read the query guide for reproducible award research.
Bundled resources
references/api-reference.md— Firecrawl-preserved Treasury API contract and endpoint index.references/query-guide.md— time/type selection, recipient resolution, and reporting.scripts/verify.pyandassets/verification-cases.json— live contract and grant checks.evals/evals.json— amount, recipient, time-window, and subaward cases.
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingdata-skills/us/usaspending-awards/SKILL.md
Popularity
- 7-day trend score
- 5
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/buriedsignals/anomaly/usaspending-awards)