US Government Data & National Debt MCP Server
Pricing
from $3.00 / 1,000 read operations
US Government Data & National Debt MCP Server
MCP server for US federal government data in Claude, ChatGPT, Cursor and VS Code. 13 read-only tools return the national debt since 1790, a federal finances snapshot, USAspending contracts and grants by agency or recipient, and Federal Register rules and executive orders since 1994.
Pricing
from $3.00 / 1,000 read operations
Rating
0.0
(0)
Developer
MrBridge
Maintained by CommunityActor stats
0
Bookmarked
1
Total users
0
Monthly active users
19 days ago
Last modified
Categories
Features
Share
🏛️ US Government Data & National Debt MCP Server
US Government Data & National Debt MCP Server is an MCP server that gives Claude, ChatGPT, Cursor, VS Code and other MCP clients three public US federal data services in one connection. Its 13 read-only tools return the national debt on any date since 1790 and a snapshot of federal finances, search federal contracts, grants and loans and rank where the money went, and find rules, proposed rules, notices and executive orders published since 1994, including documents still on public inspection. The data comes from Treasury Fiscal Data, USAspending and FederalRegister.gov, and this server is not affiliated with any of them.
Ask Claude "How big is the US national debt today, and how much has it grown over the past year?" It calls usgov_get_national_debt and answers with the latest total, the split between debt held by the public and intragovernmental holdings, and the change over one month and one year, with the dataset it came from. You need no API key and no spreadsheet.
What this server exposes
This server gives an AI client 13 read-only tools over three public US government data services. From the Treasury's Fiscal Data API it returns the national debt on any date since 1790, a snapshot of federal finances (debt, cash in the Treasury General Account, the statutory debt limit with its remaining headroom, the latest monthly and fiscal-year-to-date receipts, outlays and deficit), and rows of twelve curated tables such as average interest rates, securities auction results, interest expense, receipts by source and outlays by function, and the Treasury reporting exchange rates. From USAspending it finds federal contracts, grants, loans and direct payments, opens one award in full, lists agencies with their budgets, profiles one agency, and totals award spending by recipient, agency, industry, state or period. From the Federal Register it searches rules, proposed rules, notices and presidential documents published since 1994, reads one document with its text, and lists the documents filed for public inspection before they are published. A search and fetch pair covers Federal Register documents for ChatGPT. No account or API key is needed, and every answer names its source. This server is not affiliated with the U.S. Department of the Treasury, USAspending.gov, the Office of the Federal Register or any other federal agency.
Tools
| Tool | What it answers | Billed event |
|---|---|---|
usgov_get_national_debt | Get the US national debt | tool-read |
usgov_get_fiscal_snapshot | Get a snapshot of US federal finances | tool-action |
usgov_get_treasury_dataset | Read a curated US Treasury Fiscal Data table | tool-read |
usgov_search_awards | Search US federal awards (contracts, grants, loans) | tool-read |
usgov_get_award | Get one US federal award | tool-read |
usgov_list_agencies | List US federal agencies with their budgets | tool-read |
usgov_get_agency_profile | Get the spending profile of a US federal agency | tool-action |
usgov_get_spending_breakdown | Break down US federal award spending | tool-action |
usgov_search_federal_register | Search Federal Register documents | tool-read |
usgov_get_federal_register_document | Get one Federal Register document | tool-read |
usgov_list_public_inspection | List Federal Register documents on public inspection | tool-read |
search | Search | tool-read |
fetch | Fetch | tool-read |
usgov_get_national_debt: Get the US national debt
Use this when you need the US national debt (total public debt outstanding) today or on a past date, with its split between debt held by the public and intragovernmental holdings and its change over one month and one year, for example "How big is the national debt?" or "What was the debt in 1945?". Do not use for the debt limit or the Treasury cash balance (usgov_get_fiscal_snapshot), or for a daily series of the debt (usgov_get_treasury_dataset with dataset "debt_to_penny"). Parameters: date (optional, YYYY-MM-DD, the latest figure on or before that day, default latest), response_format ("concise" or "detailed" to add the fiscal quarter and cache information). From 1 April 1993 the figure is the daily Debt to the Penny in US dollars. Earlier dates return the fiscal year-end total of Historical Debt Outstanding (since 1790), without the split.
| Parameter | Type | Required | Example |
|---|---|---|---|
date | string | no | "2020-03-31" |
response_format | "concise" or "detailed" | no | "concise" |
Returns { as_of, source_dataset, unit, total_public_debt, debt_held_by_public, intragovernmental_holdings, fiscal_year, fiscal_quarter, changes[], notes[], attribution }. Upstream: Treasury Fiscal Data. Cache: daily (1 day). Read-only.
usgov_get_fiscal_snapshot: Get a snapshot of US federal finances
Use this when you need the current state of US federal finances in one answer: the national debt, the cash in the Treasury General Account, the statutory debt limit with the debt subject to it and the remaining headroom, and the latest monthly and fiscal-year-to-date receipts, outlays and deficit. Do not use for a past date or a time series (usgov_get_national_debt or usgov_get_treasury_dataset), or for agency budgets (usgov_get_agency_profile). Parameter: response_format ("concise" or "detailed" to add cache information). All amounts are in US dollars. Cash and debt limit figures come from the Daily Treasury Statement, rounded to the nearest million, and the headroom is computed. Deficits are positive, surpluses negative. Each part carries its own as-of date.
| Parameter | Type | Required | Example |
|---|---|---|---|
response_format | "concise" or "detailed" | no | "concise" |
Returns { unit, debt, cash, debt_limit, budget, notes[], attribution }. Upstream: Treasury Fiscal Data. Cache: intraday (5 minutes). Read-only.
usgov_get_treasury_dataset: Read a curated US Treasury Fiscal Data table
Use this when you need rows of one US Treasury Fiscal Data table over a date range: daily debt, Treasury cash deposits and withdrawals, the debt limit table, monthly receipts and outlays, receipts by source and outlays by function, average interest rates, interest expense, Treasury reporting exchange rates, securities auction results, or debt by security type. Do not use for a one-line answer about the debt or the deficit (usgov_get_national_debt, usgov_get_fiscal_snapshot), or for federal awards (usgov_search_awards). Parameters: dataset (one of "debt_to_penny" (Debt to the Penny, daily, since 1993), "historical_debt_outstanding" (Historical Debt Outstanding, annual, since 1790), "operating_cash_balance" (Operating Cash Balance, daily, since 2005, filters account_type), "deposits_withdrawals" (Deposits and Withdrawals of Operating Cash, daily, since 2005, filters transaction_type, transaction_catg), "debt_subject_to_limit" (Debt Subject to Limit, daily, since 2005, filters debt_catg), "monthly_budget_summary" (Summary of Receipts, Outlays, and the Deficit/Surplus, monthly, since 2015, filters classification_desc), "receipts_outlays_by_function" (Receipts by Source and Outlays by Function, monthly, since 2015, filters classification_desc), "average_interest_rates" (Average Interest Rates on U.S. Treasury Securities, monthly, since 2001, filters security_type_desc, security_desc), "interest_expense" (Interest Expense on the Public Debt Outstanding, monthly, since 2010, filters expense_catg_desc, expense_group_desc, expense_type_desc), "exchange_rates" (Treasury Reporting Rates of Exchange, quarterly, since 2001, filters country, currency, country_currency_desc), "auctions" (Treasury Securities Auctions Data, daily, since 1979, filters security_type, security_term, original_security_term, cusip, inflation_index_security, floating_rate), "public_debt_summary" (Summary of Treasury Securities Outstanding, monthly, since 2001, filters security_type_desc, security_class_desc)), start_date and end_date (YYYY-MM-DD, on the table date: auction date for auctions, record date otherwise), filters (object of exact, case-sensitive values on the allowed fields, for example {"security_type": "Note", "original_security_term": "10-Year", "inflation_index_security": "No"} for nominal 10-year notes, reopenings included), sort ("newest" default or "oldest"), limit (1 to 100, at most 50 in detailed mode), offset, response_format ("concise" key columns or "detailed" every curated column). Each column states its unit: some tables are in dollars, the Daily Treasury Statement and the public debt summary in millions of dollars.
| Parameter | Type | Required | Example |
|---|---|---|---|
dataset | "debt_to_penny" or "historical_debt_outstanding" or "operating_cash_balance" or "deposits_withdrawals" or "debt_subject_to_limit" or "monthly_budget_summary" or "receipts_outlays_by_function" or "average_interest_rates" or "interest_expense" or "exchange_rates" or "auctions" or "public_debt_summary" | yes | "average_interest_rates" |
start_date | string | no | "2025-01-01" |
end_date | string | no | "2025-12-31" |
filters | object | no | {"security_desc":"Treasury Notes"} |
sort | "newest" or "oldest" | no | "newest" |
limit | integer | no | 20 |
offset | integer | no | 0 |
response_format | "concise" or "detailed" | no | "concise" |
Returns { dataset, columns[], rows[], filters_applied, pagination, attribution }. Upstream: Treasury Fiscal Data. Cache: daily (1 day). Read-only.
usgov_search_awards: Search US federal awards (contracts, grants, loans)
Use this when you need individual US federal awards from USAspending: who received a contract, grant, loan or direct payment, from which agency, for how much and for what, for example "largest Department of Defense cybersecurity contracts since October 2024". Do not use for totals by recipient, agency, industry, state or period (usgov_get_spending_breakdown), agency budgets (usgov_get_agency_profile), or the full record of one award (usgov_get_award). Parameters: award_category (default "contracts"), keywords, agency, recipient, start_date and end_date (action dates from 2007-10-01), naics_codes, psc_codes, assistance_listings, state, min_amount, max_amount, award_number (the PIID, FAIN or URI of one award), sort ("amount" default, "start_date", "last_modified"), order ("desc" or "asc"), limit (1 to 50, default 10, at most 25 in detailed mode), offset (first 1,000 results), response_format. Amounts are obligated dollars (loans: face value). The award_id of each result opens the full record with usgov_get_award.
| Parameter | Type | Required | Example |
|---|---|---|---|
award_category | "contracts" or "idvs" or "grants" or "direct_payments" or "loans" or "other" | no | "contracts" |
keywords | array of string | no | ["cybersecurity"] |
agency | string | no | "DOD" |
recipient | string | no | "Lockheed Martin" |
start_date | string | no | "2024-10-01" |
end_date | string | no | "2025-09-30" |
naics_codes | array of string | no | ["541512"] |
psc_codes | array of string | no | ["D310"] |
assistance_listings | array of string | no | ["93.778"] |
state | string | no | "VA" |
min_amount | number | no | 1000000 |
max_amount | number | no | 50000000 |
award_number | string | no | "HT940216C0001" |
sort | "amount" or "start_date" or "last_modified" | no | "amount" |
order | "desc" or "asc" | no | "desc" |
limit | integer | no | 10 |
offset | integer | no | 0 |
response_format | "concise" or "detailed" | no | "concise" |
Returns { award_category, amount_meaning, total_matching, awards[], filters_applied, notes[], pagination, attribution }. Upstream: USAspending. Cache: daily (1 day). Read-only.
usgov_get_award: Get one US federal award
Use this when you need the full record of one US federal award from USAspending: type, purpose, obligated amount, ceiling and outlays (or loan value), dates, awarding and funding agencies, recipient (name, UEI, city, state, country, business categories), place of performance, industry and product codes or assistance listings, parent contract and subawards. Do not use for finding awards (usgov_search_awards) or totals (usgov_get_spending_breakdown). Parameters: award_id (the award_id returned by usgov_search_awards, for example "CONT_AWD_H907_9700_SPE2DX16D1500_9700", a numeric USAspending id, or an award page URL), response_format ("concise" or "detailed" to add competition, pricing and set-aside details and assistance listing objectives). DUNS numbers, parent company data, street addresses and executive compensation are never returned.
| Parameter | Type | Required | Example |
|---|---|---|---|
award_id | string | yes | "CONT_AWD_H907_9700_SPE2DX16D1500_9700" |
response_format | "concise" or "detailed" | no | "concise" |
Returns { award_id, award_number, category, type, description, amounts, dates, awarding_agency, funding_agency, recipient, place_of_performance, naics, psc, assistance_listings[], parent_award, subawards, competition, notes[], url, attribution }. Upstream: USAspending. Cache: daily (1 day). Read-only.
usgov_list_agencies: List US federal agencies with their budgets
Use this when you need the US federal toptier agencies with their toptier codes and abbreviations, ranked by budgetary resources, obligations or outlays of the latest reported fiscal year, for example "Which agencies have the largest budgets?" or to find the code of an agency before another call. Do not use for one agency in depth (usgov_get_agency_profile), spending by recipient or state (usgov_get_spending_breakdown), or Federal Register agencies (usgov_search_federal_register takes agency names directly). Parameters: query (optional name, abbreviation or code fragment, for example "health"), sort ("budget_authority" default, "obligations", "outlays", "name"), order ("desc" default or "asc"), limit (1 to 100, default 20), offset, response_format ("concise" or "detailed" to add the congressional budget justification link). Amounts are US dollars for the fiscal year and quarter shown on each row. The share is the agency part of all federal budgetary resources, in percent.
| Parameter | Type | Required | Example |
|---|---|---|---|
query | string | no | "health" |
sort | "budget_authority" or "obligations" or "outlays" or "name" | no | "budget_authority" |
order | "desc" or "asc" | no | "desc" |
limit | integer | no | 20 |
offset | integer | no | 0 |
response_format | "concise" or "detailed" | no | "concise" |
Returns { agencies[], pagination, attribution }. Upstream: USAspending. Cache: daily (1 day). Read-only.
usgov_get_agency_profile: Get the spending profile of a US federal agency
Use this when you need one US federal agency in depth from USAspending: mission, website, budgetary resources, obligations and outlays of a fiscal year, obligations by award type (contracts, grants, loans, direct payments), its largest sub-agencies, and in detailed mode its budget history since fiscal year 2017. Do not use for ranking all agencies (usgov_list_agencies), individual awards (usgov_search_awards), or spending by recipient, industry or state (usgov_get_spending_breakdown). Parameters: agency (toptier code "075", abbreviation "HHS" or name "Department of Health and Human Services"), fiscal_year (optional, from 2017, default the latest year the agency reported), response_format ("concise" or "detailed" to add the yearly history and the full mission text). Amounts are US dollars. Fiscal year N runs from 1 October N-1 to 30 September N.
| Parameter | Type | Required | Example |
|---|---|---|---|
agency | string | yes | "HHS" |
fiscal_year | integer | no | 2025 |
response_format | "concise" or "detailed" | no | "concise" |
Returns { agency, fiscal_year, totals, obligations_by_award_type[], total_award_obligations, top_sub_agencies, history[], notes[], attribution }. Upstream: USAspending. Cache: daily (1 day). Read-only.
usgov_get_spending_breakdown: Break down US federal award spending
Use this when you need US federal award spending totals from USAspending grouped by one dimension: top recipients, awarding agencies or sub-agencies, industries (NAICS), products and services (PSC), assistance listings, states, countries, or by fiscal year, quarter or month, for example "top recipients of NASA grants in fiscal year 2025" or "monthly Department of Energy contract obligations". Do not use for individual awards (usgov_search_awards), agency budgets (usgov_get_agency_profile), or Treasury receipts and outlays (usgov_get_treasury_dataset). Parameters: group_by (recipient, awarding_agency, awarding_subagency, naics, psc, assistance_listing, state, country, fiscal_year, quarter, month), optional filters award_category, keywords, agency, recipient, start_date, end_date (action dates from 2007-10-01), naics_codes, psc_codes, assistance_listings, state, min_amount, max_amount, then limit (1 to 100, default 10), offset (rankings up to 1,000), response_format ("detailed" adds outlays and, for periods, the split by award type). Amounts are obligations in US dollars of the transactions dated in the period. Rankings are largest first, periods newest first, and a fiscal year starts in October (quarter 1 and month 1 are October).
| Parameter | Type | Required | Example |
|---|---|---|---|
group_by | "recipient" or "awarding_agency" or "awarding_subagency" or "naics" or "psc" or "assistance_listing" or "state" or "country" or "fiscal_year" or "quarter" or "month" | yes | "recipient" |
award_category | "contracts" or "idvs" or "grants" or "direct_payments" or "loans" or "other" | no | "grants" |
keywords | array of string | no | ["cybersecurity"] |
agency | string | no | "DOD" |
recipient | string | no | "Lockheed Martin" |
start_date | string | no | "2024-10-01" |
end_date | string | no | "2025-09-30" |
naics_codes | array of string | no | ["541512"] |
psc_codes | array of string | no | ["D310"] |
assistance_listings | array of string | no | ["93.778"] |
state | string | no | "VA" |
min_amount | number | no | 1000000 |
max_amount | number | no | 50000000 |
limit | integer | no | 10 |
offset | integer | no | 0 |
response_format | "concise" or "detailed" | no | "concise" |
Returns { group_by, rows[], filters_applied, total_is_exact, notes[], pagination, attribution }. Upstream: USAspending. Cache: daily (1 day). Read-only.
usgov_search_federal_register: Search Federal Register documents
Use this when you need US Federal Register documents published since 1994: final rules, proposed rules, notices and presidential documents (executive orders, proclamations, memoranda), by words, agency, type, date or docket, for example "EPA proposed rules on methane in 2025" or "executive orders signed this year". Do not use for documents not yet published (usgov_list_public_inspection), the full text of one document (usgov_get_federal_register_document), or federal spending (usgov_search_awards). Parameters: query (words, a phrase in quotes, or a citation such as "89 FR 12345"), document_types (rule, proposed_rule, notice, presidential_document), presidential_document_types (executive_order, proclamation, memorandum, determination, notice, presidential_order, other), agencies (names, short names such as "EPA", or slugs), start_date and end_date (publication dates), significant (true for significant regulatory actions under Executive Order 12866), docket_id, order ("relevance" default with a query, else "newest", or "oldest"), limit (1 to 50, default 10, at most 25 in detailed mode), offset (first 2,000 results), response_format.
| Parameter | Type | Required | Example |
|---|---|---|---|
query | string | no | "artificial intelligence" |
document_types | array of string | no | ["proposed_rule"] |
presidential_document_types | array of string | no | ["executive_order"] |
agencies | array of string | no | ["EPA"] |
start_date | string | no | "2025-01-01" |
end_date | string | no | "2025-12-31" |
significant | boolean | no | true |
docket_id | string | no | "EPA-HQ-OAR-2021-0317" |
order | "relevance" or "newest" or "oldest" | no | "newest" |
limit | integer | no | 10 |
offset | integer | no | 0 |
response_format | "concise" or "detailed" | no | "concise" |
Returns { description, documents[], notes[], pagination, attribution }. Upstream: Federal Register. Cache: intraday (5 minutes). Read-only.
usgov_get_federal_register_document: Get one Federal Register document
Use this when you need one published US Federal Register document: its title, type, agencies, abstract, action, key dates (effective, comments close, signing), citation, dockets and CFR parts affected (first 20 of each, with their totals), links to the page and the official PDF, and optionally its plain text. Do not use for finding documents (usgov_search_federal_register) or documents not yet published (usgov_list_public_inspection). Parameters: document_number (for example "2025-02095", or a federalregister.gov document URL), include_full_text (default false, true reads the plain text of the document), max_text_chars (1,000 to 40,000, default 12,000), response_format ("concise" or "detailed" to add topics, RINs, pages, corrections and the regulations.gov link). The text comes from FederalRegister.gov, which is not the official legal edition: pdf_url is the official version on govinfo.gov.
| Parameter | Type | Required | Example |
|---|---|---|---|
document_number | string | yes | "2025-02095" |
include_full_text | boolean | no | false |
max_text_chars | integer | no | 12000 |
response_format | "concise" or "detailed" | no | "concise" |
Returns { document_number, title, type, subtype, action, abstract, agencies[], publication_date, citation, dates, effective_on, comments_close_on, signing_date, executive_order_number, docket_ids[], docket_ids_total, cfr_references[], cfr_references_total, significant, comment_url, html_url, pdf_url, president, topics[], topics_total, regulation_id_numbers[], regulation_id_numbers_total, pages, correction_of, corrections[], corrections_total, regulations_dot_gov_url, public_inspection_pdf_url, full_text, attribution }. Upstream: Federal Register, Federal Register text. Cache: daily (1 day). Read-only.
usgov_list_public_inspection: List Federal Register documents on public inspection
Use this when you need the documents filed at the Office of the Federal Register for public inspection, which are published in the next issues: upcoming rules, proposed rules, notices and presidential documents, for example "What will the Federal Register publish tomorrow?" or "Has the EPA filed anything today?". Do not use for published documents (usgov_search_federal_register). Parameters: date (optional, YYYY-MM-DD issue of the public inspection desk since 2011-11-01, default the current issue), agencies (names, short names or slugs), document_types (rule, proposed_rule, notice, presidential_document), filing_type ("regular" or "special"), query (words that must all appear in the title), limit (1 to 50, default 20), offset, response_format ("detailed" adds dockets and editorial notes). The desk opens around 8:45 a.m. Eastern time on federal business days and is empty overnight.
| Parameter | Type | Required | Example |
|---|---|---|---|
date | string | no | "2026-09-25" |
agencies | array of string | no | ["EPA"] |
document_types | array of string | no | ["rule"] |
filing_type | "regular" or "special" | no | "special" |
query | string | no | "emissions" |
limit | integer | no | 20 |
offset | integer | no | 0 |
response_format | "concise" or "detailed" | no | "concise" |
Returns { issue, documents[], pagination, attribution }. Upstream: Federal Register. Cache: intraday (5 minutes). Read-only.
search: Search
Use this when ChatGPT needs to find US Federal Register documents (rules, proposed rules, notices, executive orders and other presidential documents published since 1994) by words, a phrase or a citation such as "89 FR 12345". Do not use for federal spending or Treasury data, which have their own usgov_ tools. Parameter query (string), for example "artificial intelligence executive order". Returns up to 10 documents, most relevant first.
| Parameter | Type | Required | Example |
|---|---|---|---|
query | string | yes | "methane emissions standards" |
Returns { results[], notices[] }. Upstream: Federal Register, Federal Register text. Cache: intraday (5 minutes). Read-only.
fetch: Fetch
Use this when ChatGPT needs the content of one Federal Register document returned by search: title, type, agencies, dates, abstract and up to 20,000 characters of its text, with the link to the official PDF. Do not use for ids that did not come from search. Parameter id (document number from search), for example "2025-02095".
| Parameter | Type | Required | Example |
|---|---|---|---|
id | string | yes | "2025-02095" |
Returns { id, title, text, url, metadata, notices[] }. Upstream: Federal Register, Federal Register text. Cache: intraday (5 minutes). Read-only.
Connect
Endpoint (Streamable HTTP): https://mrbridge--us-gov-data-mcp.apify.actor/mcp
Every client authenticates with your Apify API token, found in Apify Console > Settings > API & Integrations. Clients that support custom headers send it as Authorization: Bearer <token>. The others use the ?token= URL variant, which must be treated as a secret: keep it out of commits, screenshots and shared logs.
Claude Code
$claude mcp add --transport http us-gov-data "https://mrbridge--us-gov-data-mcp.apify.actor/mcp" --header "Authorization: Bearer ${APIFY_TOKEN}"
Cursor, VS Code and other JSON configurations
{"mcpServers": {"us-gov-data": {"type": "http","url": "https://mrbridge--us-gov-data-mcp.apify.actor/mcp","headers": {"Authorization": "Bearer ${APIFY_TOKEN}"}}}}
claude.ai and Claude Desktop
Add a custom connector with the URL https://mrbridge--us-gov-data-mcp.apify.actor/mcp?token=YOUR_APIFY_TOKEN and no authentication. Prefilled link (replace the token placeholder before saving): add US Government Data & National Debt MCP Server to Claude.
ChatGPT (developer mode)
Settings > Apps > Create, URL https://mrbridge--us-gov-data-mcp.apify.actor/mcp?token=YOUR_APIFY_TOKEN, authentication "No authentication". ChatGPT snapshots the tool list: after a server update, an admin has to click Refresh on the app.
OpenAI Responses API
{"type": "mcp","server_label": "us_gov_data","server_url": "https://mrbridge--us-gov-data-mcp.apify.actor/mcp","headers": {"Authorization": "Bearer YOUR_APIFY_TOKEN"},"require_approval": "never"}
Add allowed_tools to restrict the tool list. Tool calls must finish within about 60 seconds on this API, and every tool of this server answers in under 50 seconds.
Anthropic Messages API
{"mcp_servers": [{"type": "url","url": "https://mrbridge--us-gov-data-mcp.apify.actor/mcp","name": "us-gov-data","authorization_token": "YOUR_APIFY_TOKEN"}],"tools": [{"type": "mcp_toolset","mcp_server_name": "us-gov-data"}]}
Send the beta header anthropic-beta: mcp-client-2025-11-20.
One-click OAuth alternative
https://mcp.apify.com/?tools=mrbridge/us-gov-data-mcp connects through Apify OAuth without a token. It exposes the Actor as a single tool rather than the granular tools listed above.
Cost control
Each Standby run has a spending limit (maxTotalChargeUsd). When a call uses the last of it, the result is still returned, and later calls in that run answer CHARGE_LIMIT_REACHED until the limit is raised.
Why use US Government Data & National Debt MCP Server?
Federal fiscal, spending and regulatory data sit on three different APIs, each with its own filters and paging rules. This server hides those differences behind tools an assistant can call from a plain question.
- Headline answers in one call.
usgov_get_fiscal_snapshotreturns the debt, the cash in the Treasury General Account, the debt limit with its remaining headroom and the latest deficit, computed the way the published statement does it. - Agencies by any name. Pass a code, an abbreviation such as
DODorHHS, or a full name, and the server resolves it before searching awards or budgets. - Spending rankings, not just lists. Total award spending by recipient, agency, industry code, product code, state or month, with exact totals for award searches.
- Rules before they are published.
usgov_list_public_inspectionlists the documents filed for the next issues of the Federal Register. - ChatGPT deep research ready. A
searchandfetchpair covers Federal Register documents, with up to 20,000 characters of text and the link to the PDF on govinfo.gov. - 🔌 Model Context Protocol over Streamable HTTP at a single endpoint,
https://mrbridge--us-gov-data-mcp.apify.actor/mcp, for Claude, ChatGPT, Cursor, VS Code and the OpenAI and Anthropic APIs. - ⚡ Apify Standby mode. The first call after a quiet spell starts the server in about 5 seconds (measured on the platform on 2026-10-05). Later calls skip that wait until the server has been idle for 5 minutes.
- Sources stated in every answer. Each answer carries an attribution block with the provider, the dataset and its terms page, plus the Dun & Bradstreet credit where recipient names appear.
Pricing (Pay-per-Event)
| Event | Charged when | Price |
|---|---|---|
apify-actor-start | run start (platform default) | see the Store page |
tool-read (primary event) | each successful read tool call | see the Store page |
tool-action | each successful call of a detailed or computed tool | see the Store page |
Never charged: initialize, tools/list, pings, readiness probes, and any call that ends in an error (invalid input, not found, upstream failure, timeout, spending limit).
Platform usage comes on top of these events. This server runs in Apify Standby mode, and you also pay the platform usage of your own Standby run, billed to your Apify account like any other run: the memory set for the Standby run, for as long as the server is up, idle time included, until the run stops at its idle timeout. The Apify pay-per-event documentation says so: "For runs in Standby mode, users always pay the platform usage costs in addition to the event charges."
Limits, caveats and attribution
- Attribution. Every answer carries an
attributionblock: the provider, the dataset or service, the terms URL and the notices below. Keep it when you reuse the figures. - Treasury Fiscal Data. Twelve curated tables, never a free query. Filter values must match the stored spelling (the server corrects the case of known values such as "Note" or "10-Year"), and values containing a comma or a colon cannot be filtered. Units differ by table: the Daily Treasury Statement and the public debt summary are in millions of dollars, the other tables in dollars, and each column states its unit. The fiscal snapshot converts everything to dollars and computes the debt subject to limit and the headroom the way the published statement does. Before 1 April 1993 only fiscal year-end totals exist, without the split between the public and intragovernmental holdings.
- USAspending. Searches start on 1 October 2007 (fiscal year 2008). Award amounts are obligated dollars, not contract ceilings, and loan obligations are subsidy costs. Department of Defense and Army Corps of Engineers contract data appear with a 90-day delay. A search pages through its first 1,000 results and a ranking through its top 1,000, at most 50 awards or 100 rows per call. Payments to individuals appear as USAspending publishes them ("REDACTED DUE TO PII", "MULTIPLE RECIPIENTS"). USAspending slows down clients that send more than about 1,000 requests in 5 minutes, so heavy use can meet a temporary
UPSTREAM_RATE_LIMITED. - Dun & Bradstreet. Recipient names and locations, mostly in awards made before April 2022, may contain D&B Open Data. Every answer that shows them credits Dun & Bradstreet, Inc. This server never returns DUNS numbers, parent company data, street addresses or executive compensation, and never stores them either: they are removed from every USAspending answer before its cache and its logs.
- Federal Register. FederalRegister.gov is not the official legal edition of the Federal Register and gives no legal notice. Each document links to its official PDF on govinfo.gov: rely on that version for legal research. A search pages through its first 2,000 results (API limit), narrow the dates to reach more. The document text is read when asked, cut at 40,000 characters, and when the FederalRegister.gov website refuses the download the answer keeps the metadata and says why. Long lists (dockets, CFR parts, topics) show their first 20 items and their total. The server keeps well under the 1,000 requests per hour FederalRegister.gov asks for, so a busy hour can meet a temporary
UPSTREAM_RATE_LIMITED. Documents on public inspection are not yet published. - Freshness. Latest Treasury figures (and dates of the last 3 days), Federal Register searches and recent public inspection issues are cached for 5 minutes, older Treasury history, USAspending answers and Federal Register documents for one day, agency lists and document texts for 30 days.
Data terms
By using this server you agree to use the data it returns in line with the terms of each source, and to pass the same obligations on to anyone you share the data with.
- Treasury Fiscal Data: API documentation, License and Authorization. The data is free to copy, adapt and redistribute, commercial use included. Do not suggest Treasury endorsement.
- USAspending: About USAspending and the Limitation on Permissible Use of Dun & Bradstreet, Inc. Data. Credit Dun & Bradstreet for recipient names and addresses, do not use or share them in bulk, and do not use any other D&B data for commercial, resale or marketing purposes.
- Federal Register: Developer resources and About this site. The text may be reproduced freely (1 CFR 2.6), but material quoted from third parties keeps its own rights, NARA and OFR logos or seals may not be used, and only the official edition on govinfo.gov gives legal notice.
The fee for this server pays for the service (hosting, caching, normalisation), not for a licence on the data.
Use cases
Debt and deficit fact checks. A reporter asks for the national debt on 31 March 2020 and the deficit of the latest fiscal year to date. usgov_get_national_debt with date: "2020-03-31" and usgov_get_fiscal_snapshot answer both, each with its dataset and date.
Debt limit watch. "How much room is left under the debt limit, and how much cash does the Treasury hold?" The fiscal snapshot answers in one call.
Contract research. "Show the largest Department of Defense cybersecurity contracts since October 2024." usgov_search_awards filters on the agency, the keyword and the start date, sorts by amount, then usgov_get_award opens any of them in full, down to the place of performance and the industry code.
Grant landscape by state. "Which states received the most HHS grant money in fiscal year 2025?" usgov_get_spending_breakdown groups grants by state over the fiscal year window.
Regulatory monitoring. "List EPA proposed rules on methane from 2025, then tell me what the EPA filed for public inspection today." The assistant combines usgov_search_federal_register and usgov_list_public_inspection, and reads a document with usgov_get_federal_register_document when you want the comment deadline or the CFR parts affected.
Agency budgets. usgov_list_agencies ranks federal agencies by budgetary resources, and usgov_get_agency_profile breaks one down by award type and sub-agency, with its budget history since fiscal year 2017 in detailed mode.
More company and economic data tools
- LEI search MCP server: legal entity identifiers by name or register number, parents and subsidiaries, ISIN and BIC lookups.
- US company filings MCP server: 10-K, 10-Q and 8-K filings and reported figures over time, read from SEC EDGAR data.
- economic indicators MCP server: GDP, inflation and debt for about 200 countries, Eurostat statistics and ECB exchange rates.
- UK public data for AI: police-recorded crime near a point, MPs, bills and Commons votes from UK open data.
Frequently asked questions
Do I need an API key or a government account?
No. The three data services are public and keyless, and this server needs no contact e-mail. You only need an Apify account and its API token.
How much does it cost to query US government data?
Two things are billed to your Apify account. First, the tool calls: $0.01 per successful call to usgov_get_fiscal_snapshot, usgov_get_agency_profile or usgov_get_spending_breakdown (actions) and $0.005 per successful call to any other tool (reads). These are the prices on the Apify Free plan, and the discounts of paid plans lower them. Errors, invalid input, the spending-limit refusal and protocol traffic (initialize, tool lists, pings) are never charged. Second, the platform usage of your own Standby run, on top of the events: the memory set for the run (256 MB) for as long as the server is up, idle time included, until it stops 5 minutes after your last call, plus a small start event per new run. A 10-minute session of 30 calls, 6 of them actions, comes to $0.18 in tool calls on the Free plan plus 1 to 2 cents of platform usage. Live prices for your Apify plan are on the Pricing tab, and maxTotalChargeUsd caps the event charges of one run.
What periods and records are covered?
The national debt goes back to 1790, with fiscal year-end totals only before 1 April 1993. Award searches start on 1 October 2007, and Department of Defense and Army Corps of Engineers contracts appear with a 90-day delay. Federal Register documents go back to 1994. A search pages through its first 1,000 awards or its top 1,000 ranking rows, and through the first 2,000 Federal Register results: narrow the dates to reach more.
How fresh is the data?
The latest Treasury figures, Federal Register searches and recent public inspection issues are cached for 5 minutes. Older Treasury history, award data and Federal Register documents are cached for a day, agency lists and document texts for 30 days. The sources publish on their own schedules, from the next business day for the daily debt to monthly for the budget statement.
Can I export every award or build a contractor list?
No. Recipient names and locations in older awards may contain Dun & Bradstreet data, whose terms forbid bulk use. Each call returns at most 50 awards or 100 rows, there is no export tool, and DUNS numbers, parent company data, street addresses and executive compensation are never returned.
Can I rely on these Federal Register documents for legal work?
Use them to find and read documents, then check the official PDF on govinfo.gov that comes with each one. FederalRegister.gov is not the official legal edition of the Federal Register and gives no legal notice, and nothing this server returns is legal advice.
Is this server affiliated with the US government?
No. It is an independent server that reads public data and names its source in every answer. It is not affiliated with or endorsed by the U.S. Department of the Treasury, USAspending.gov, the Office of the Federal Register, NARA, GPO or any other federal agency, and it uses none of their seals or logos.
Why is my first call slow?
Cold start. After 5 idle minutes the server stops, and the next call waits about 5 seconds while Apify starts a new run.
Your feedback
Found a wrong figure or a broken tool? Open an issue on the Issues tab and it will be looked at. More MCP servers, scrapers and data studies from the same team are on mr-bridge.com.
Changelog
- 0.1.5: Output, dataset and key-value store schemas, as Apify asks for publication: the Output tab of a run started from the Console shows the connection details and one row per tool.
- 0.1.4: the Input tab shows the endpoint and how to connect an AI client, and a run started from the Console outputs the connection details and the tool list
- 0.1.3: pricing events tool-read and tool-action.
usgov_get_fiscal_snapshot,usgov_get_agency_profileandusgov_get_spending_breakdownbill onetool-actionevent, every other tool onetool-readevent - 0.1.2: server name aligned with the Store title
- 0.1.1: the README intro and the server instructions call the three sources public instead of official, as the source terms ask (no wording that presents a source as a partner).
- 0.1.0: initial build (private, unpublished)