Skip to content

usaspending-mcp-server

io.github.cyanheads/usaspending-mcp-servercyanheadsstreamable-httpv0.7.3Scanned

Access US federal award, recipient, agency, and spending analytics data from USAspending.gov.

18 tools: 3 change things and 15 only read.

Tools: 0 delete, 0 irreversible, 3 write, 15 read.
  • Reads private data · 1
  • Reads text others write · 8
  • Can send data out · 0
  • No findings

Tools listed without sign-in. Remote posture

With Anlyon

3 tools here change things. Govern them.

  • Anlyon lists the tools and adopts none.
  • Each tool you adopt gets approvals and receipts.
  • Your agent never holds the key.
Govern this server →

A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.

What the usaspending-mcp-server can do

The usaspending-mcp-server is published by the GitHub account cyanheads. It has 18 tools. It can create or change data (3 tools: usaspending_autocomplete_filters, usaspending_spending_by_geography and usaspending_spending_by_category), but cannot delete data. 15 tools only read.

1 tool reads private data (usaspending_search_federal_accounts); 8 tools bring in text other people can write (usaspending_autocomplete_filters, usaspending_search_awards, usaspending_search_recipients and 5 more).

None of the fixed checks matched anything in its tool definitions. That is not a safety guarantee.

Its remote endpoint lists its tools without sign-in, and its tools changed 5 times since Anlyon Radar first saw them.

Last scanned Nov 14, 2026, version 0.7.3.

Tools

Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.

18 tools

Changes things 3Claude Code: ask
  • usaspending_autocomplete_filtersLook up valid code values for filter fields by searching free-text descriptions.Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Look up valid code values for filter fields by searching free-text descriptions. Use the type parameter to select the lookup table: naics (NAICS industry codes), psc (product/service codes), cfda (CFDA/Assistance Listing program numbers), def_code (Disaster/Emergency Funding codes by code, appropriation title, or public law), awarding_agency (agency and sub-agency names, each with its tier and top-tier parent — agency_name and funding_agency_name filters take the matched name; a sub-tier name shared by several parents matches under all parents), recipient (recipient names with UEI/DUNS), or location (US state, county, and congressional district names, resolved to the codes the location filters and usaspending_spending_by_geography shape_codes take). Call this before filtering awards when you know a description but not the exact code. Returns matching codes and names for use in other tool filters.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • usaspending_spending_by_categoryAggregate federal spending grouped by a specific dimension: NAICS industry code, PSC product/service code, awarding or funding agency or subagency, CFDA assistance program, recipie…Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Aggregate federal spending grouped by a specific dimension: NAICS industry code, PSC product/service code, awarding or funding agency or subagency, CFDA assistance program, recipient, federal account, or Disaster Emergency Fund Code (DEF). Returns top items with obligation amounts — useful for trend and breakdown analysis. Filter by keyword, award type, agency, recipient, NAICS code, Assistance Listing, PSC, DEF code, place of performance or recipient location, or date range. federal_account and defc amounts are whole transaction obligations, counted under every federal account or DEF code the award carries, so rows overlap and never sum to a total; on defc, filters.def_codes selects awards but the rows still list every code those awards carry. Chain NAICS, PSC, CFDA, and DEF codes back into these filters, into usaspending_search_awards filters, or into usaspending_autocomplete_filters lookups; a federal_account code (AGENCY-MAIN, e.g. 047-4534) opens in usaspending_get_federal_account. Agency rows carry agency_slug, which opens the agency in usaspending_get_agency — on a subagency row, its parent department. recipient_duns rows carry recipient_id, which opens the recipient profile in usaspending_get_recipient, and uei.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • usaspending_spending_by_geographyAggregate federal spending by state, county, or congressional district.Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Aggregate federal spending by state, county, or congressional district. Useful for per-capita analysis, regional comparisons, and mapping federal investment patterns. Filter by keyword, award type, agency, recipient, NAICS code, Assistance Listing, PSC, DEF code, place of performance or recipient location, or date range. scope picks the location results are grouped by, separately from the location filters: the counties of one state are geo_layer county with filters.place_of_performance_location set to that state and scope place_of_performance. Areas are ranked by obligation and capped at limit; to retrieve specific areas, including ones ranked past the cap, list their codes in shape_codes — a 2-letter postal code for state, a 5-digit state + county FIPS code for county, or a 4-digit state FIPS + district number for district — NOT place names. Resolve a state, county, or district name to its code with usaspending_autocomplete_filters type=location. Chain per-capita results with Census population data for meaningful comparisons.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
Only reads 15Show 15 read-only tools
  • usaspending_disaster_spendingFetch disaster and emergency supplemental spending broken down by agency, CFDA assistance program, recipient, or geography, or request an overview.Reads text others writeNo findings

    Why it is in this group

    • The description's first verb is "fetch": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Fetch disaster and emergency supplemental spending broken down by agency, CFDA assistance program, recipient, or geography, or request an overview. Find DEF codes in usaspending_autocomplete_filters with type def_code. Filter by those codes to select emergency appropriations, then check amount_coverage before interpreting amounts: award breakdowns and geography cover the provider COVID grouping, while overview and agency total cover all requested codes. Rows and award counts can match codes outside the monetary scope.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • usaspending_get_agencyFetch an agency's fiscal-year overview including mission, budgetary resources, obligation and outlay totals (for the most recent fiscal year with reported budget data), sub-agency …No findings

    Why it is in this group

    • The name "usaspending_get_agency" says "get": a word for reading.
    Description, as the server wrote it
    Fetch an agency's fiscal-year overview including mission, budgetary resources, obligation and outlay totals (for the most recent fiscal year with reported budget data), sub-agency count, and disaster/emergency funding codes. def_codes_fiscal_year names the fiscal year covered by def_codes, which follows the budget totals when available. budget_by_fiscal_year carries the same totals for every fiscal year with reported data since FY2017, newest first, for budget trends; a row whose reported_through_period is below 12 is year-to-date, so compare it to full years with care. Also returns a paginated sub-agency breakdown with obligation and transaction counts for the same fiscal year as the budget totals (the current fiscal year when no budget year is available within 5 seconds); sub_agency_fiscal_year names the year it covers. Accepts exactly one directory toptier_code (e.g., 097 for DoD, 1601 for Labor; one/two-digit numeric inputs are zero-padded to three digits) or an agency_slug (e.g., department-of-defense) — both appear in usaspending_list_agencies results and award search results. When the budget, sub-agency or DEF code request fails, the overview is still returned and unavailable_sections names each missing section with its reason.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: false
    • idempotentHint: true
  • usaspending_get_awardFetch full details of a federal award by its generated unique award ID.No findings

    Why it is in this group

    • The name "usaspending_get_award" says "get": a word for reading.
    Description, as the server wrote it
    Fetch full details of a federal award by its generated unique award ID. Returns contract or assistance award data including recipient identity (UEI, parent organization, business categories), agency hierarchy, period of performance, place of performance with its county code, Assistance Listings with the obligation under each, loan face value and subsidy cost for loans, funding account linkages (account_obligations_by_defc), parent IDV information, and subaward count and total. For contracts and IDVs it also returns competition (extent competed, solicitation procedure, set-aside, the authority for not competing, the fair-opportunity status of an order under a multiple-award IDV, offers received) and contract_pricing. Use generated_internal_id values from usaspending_search_awards as input. Recipient and parent recipient hashes can be passed to usaspending_get_recipient; NAICS codes can be used in usaspending_search_awards filters, Assistance Listing numbers in its assistance_listings filter, and the place of performance county_code with state_code in its location_filter. For IDV-category awards (category="idv"), use usaspending_get_idv_awards to list the child contracts and task/delivery orders placed under them.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: false
    • idempotentHint: true
  • usaspending_get_award_federal_accountsList the Treasury federal accounts that funded an award, with the amount obligated from each and the funding agency behind it.No findings

    Why it is in this group

    • The name "usaspending_get_award_federal_accounts" says "get": a word for reading.
    Description, as the server wrote it
    List the Treasury federal accounts that funded an award, with the amount obligated from each and the funding agency behind it. This is the award → appropriation link: each row returns federal_account (AGENCY-MAIN format, e.g. 080-0120) to chain into usaspending_get_federal_account for the account budget detail. The award_id must be a generated_unique_award_id — from usaspending_search_awards (generated_internal_id field) or usaspending_get_award. Distinct from usaspending_get_award account_obligations_by_defc, which breaks funding down by Disaster/Emergency Funding code rather than by account. An award_id that does not exist returns an empty list rather than an error.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: false
    • idempotentHint: true
  • usaspending_get_award_subawardsList subaward contracts or grants under a prime federal award.No findings

    Why it is in this group

    • The name "usaspending_get_award_subawards" says "get": a word for reading.
    Description, as the server wrote it
    List subaward contracts or grants under a prime federal award. Reveals the sub-contractor or sub-grantee layer — the organizations that actually perform the work. Each row shows the subaward number, amount, description, action date, and recipient. Check subaward_count on usaspending_get_award first to confirm subawards exist before calling this tool.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: false
    • idempotentHint: true
  • usaspending_get_award_transactionsList individual transactions (contract modifications, grant amendments) on a federal award.No findings

    Why it is in this group

    • The name "usaspending_get_award_transactions" says "get": a word for reading.
    Description, as the server wrote it
    List individual transactions (contract modifications, grant amendments) on a federal award. Each transaction represents a change event — obligation modifications, performance period extensions, scope changes, etc. Use this to trace the spending history and obligation changes over the life of an award. Award IDs come from usaspending_search_awards (generated_internal_id field).

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: false
    • idempotentHint: true
  • usaspending_get_federal_accountFetch a federal account's budget data: total obligations, gross outlays, and budgetary resources, plus the per-Treasury-Account-Symbol (TAS) component breakdown in children.No findings

    Why it is in this group

    • The name "usaspending_get_federal_account" says "get": a word for reading.
    Description, as the server wrote it
    Fetch a federal account's budget data: total obligations, gross outlays, and budgetary resources, plus the per-Treasury-Account-Symbol (TAS) component breakdown in children. fiscal_year selects a reporting year; omit it for the current federal fiscal year. An empty year retains metadata and a notice, without substituting another year. Federal accounts connect appropriations law to actual agency spending. Account codes come from usaspending_search_federal_accounts (its account_number output field) or usaspending_get_award_federal_accounts (its federal_account field), and are formatted as AGENCY-MAIN (e.g., 097-0100 for DoD Operation and Maintenance). For obligations broken down by program activity or object class, use usaspending_get_federal_account_breakdown.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: false
    • idempotentHint: true
  • usaspending_get_federal_account_breakdownFetch a federal account's obligations broken down by program activity (what the money funds) or object class (what it buys — personnel, supplies, contracts).No findings

    Why it is in this group

    • The name "usaspending_get_federal_account_breakdown" says "get": a word for reading.
    Description, as the server wrote it
    Fetch a federal account's obligations broken down by program activity (what the money funds) or object class (what it buys — personnel, supplies, contracts). Use the dimension parameter to select the axis. fiscal_year scopes financial reporting periods to the full October–September year, not award action dates; omit it for all available periods. Account codes are AGENCY-MAIN format and come from usaspending_search_federal_accounts (its account_number output field), usaspending_get_award_federal_accounts (its federal_account field), or usaspending_get_federal_account. Paginated with an honest total count. For the account's own metadata and top-level totals, use usaspending_get_federal_account.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: false
    • idempotentHint: true
  • usaspending_get_idv_awardsList child contracts and task/delivery orders placed under an IDV (Indefinite Delivery Vehicle) award.No findings

    Why it is in this group

    • The name "usaspending_get_idv_awards" says "get": a word for reading.
    Description, as the server wrote it
    List child contracts and task/delivery orders placed under an IDV (Indefinite Delivery Vehicle) award. Each row includes the generated_unique_award_id to chain into usaspending_get_award for full detail. The award_id must be the generated_unique_award_id of the parent IDV — obtainable from usaspending_search_awards (generated_internal_id field) or from usaspending_get_award. IDV category awards returned by usaspending_get_award have child orders accessible via this tool.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: false
    • idempotentHint: true
  • usaspending_get_recipientFetch a recipient's full profile including address, business type codes, parent organizations, alternate names, and total transaction and loan amounts.No findings

    Why it is in this group

    • The name "usaspending_get_recipient" says "get": a word for reading.
    Description, as the server wrote it
    Fetch a recipient's full profile including address, business type codes, parent organizations, alternate names, and total transaction and loan amounts. A -P ID covers awards USAspending attributes to that parent, and historical attribution can vary. A -C ID covers the entity's own awards recorded with a parent, and a -R ID covers its own awards recorded without one. Use IDs returned by usaspending_search_recipients or usaspending_get_award. If search returns both -C and -R for the entity, check both. Never construct an ID by replacing a suffix. Totals cover the trailing 12 months by default. Set fiscal_year to one year or "all" for USAspending's searchable history through the request date. The period field names the window. parents lists every parent organization — a child often has several, and a parent lists itself — and each parent_id opens that parent in usaspending_get_recipient. For a -P parent, children lists its child recipients with their amounts over the same period, highest amount first, up to children_limit, and children_total counts them all; children is a best-effort second request, so when it fails the profile is still returned and unavailable_sections names children with its reason. For a recipient's totals by award type, use usaspending_spending_over_time with filters.recipient_id and award_type_codes. UEI and DUNS values can be used to cross-reference with SAM.gov and SEC EDGAR.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: false
    • idempotentHint: true
  • usaspending_list_agenciesList top-tier federal agencies with toptier codes, agency slugs, budget authority amounts, and obligation totals for the latest reported period.No findings

    Why it is in this group

    • The name "usaspending_list_agencies" says "list": a word for reading.
    Description, as the server wrote it
    List top-tier federal agencies with toptier codes, agency slugs, budget authority amounts, and obligation totals for the latest reported period. Returns the full directory unless limit is supplied; optional pagination slices the provider-sorted directory. Use this as the entry point for agency navigation — toptier codes and agency slugs are required inputs for usaspending_get_agency and agency-based filters on spending analysis tools.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: false
    • idempotentHint: true
  • usaspending_search_awardsSearch federal awards by keyword, recipient, agency, award type, NAICS code, assistance listing (CFDA) number, PSC, DEF code, location, or date range.Reads text others writeNo findings

    Why it is in this group

    • The name "usaspending_search_awards" says "search": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Search federal awards by keyword, recipient, agency, award type, NAICS code, assistance listing (CFDA) number, PSC, DEF code, location, or date range. Location filters by place of performance (location_filter, or filters.place_of_performance_location, which overrides it) or by where the recipient is based (filters.recipient_location). Returns ranked award summaries including recipient names, amounts, awarding agencies, and generated award IDs for use with usaspending_get_award; loan rows carry loan value, subsidy cost, and issue date in place of award amount and dates. Rows carry the recipient_id that opens usaspending_get_recipient, and the recipient UEI, when upstream has them; page_metadata.total counts every match in the selected award-type group. Award types: A/B/C/D = contracts, IDV_A–IDV_E = IDVs, 02/03/04/05/F001/F002 = grants, 06/10/F006/F007 = direct payments, 07/08/F003/F004 = loans, 09/11/-1/F005/F008/F009/F010 = other assistance. Dates must be ISO 8601 (YYYY-MM-DD). Earliest data: 2007-10-01 via search API. DoD contracts have a 90-day publication lag.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • usaspending_search_federal_accountsList and keyword-search federal accounts by directory toptier code or title keyword.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "usaspending_search_federal_accounts" says "search": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "list" and "directory": it reads files.
    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    List and keyword-search federal accounts by directory toptier code or title keyword. Returns account numbers, names, managing agencies, and budgetary resources. fiscal_year selects the resource amounts, not account membership; omit it for the latest available submission year. Search and detail totals use different calculations and can differ for the same year. Use account_number from results as input to usaspending_get_federal_account for full budget detail. Use usaspending_list_agencies to look up the input agency_identifier (directory toptier_code, e.g. "097" for DoD or "1601" for Labor); account rows return their account-prefix agency_identifier, e.g. "016" for Labor.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • usaspending_search_recipientsSearch for organizations or individuals receiving federal funds by name, UEI (Unique Entity Identifier), or DUNS.Reads text others writeNo findings

    Why it is in this group

    • The name "usaspending_search_recipients" says "search": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Search for organizations or individuals receiving federal funds by name, UEI (Unique Entity Identifier), or DUNS. Returns recipient hash IDs, UEI/DUNS identifiers, trailing-12-month obligations, and hierarchy level. Amounts come from USAspending's precomputed snapshot and may differ from trailing-12-month profile totals. Results are paginated — use page to retrieve matches beyond the first page; page_metadata.total reports the full match count. Recipient hash IDs from this tool can be passed to usaspending_get_recipient for full profiles. Recipient level: P = parent organization, C = child entity, R = standalone.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • usaspending_spending_over_timeFetch aggregated federal obligation amounts grouped by fiscal year, fiscal quarter, or fiscal month.Reads text others writeNo findings

    Why it is in this group

    • The description's first verb is "fetch": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Fetch aggregated federal obligation amounts grouped by fiscal year, fiscal quarter, or fiscal month. All grouping is relative to the US government fiscal year (Oct–Sep), so fiscal month 1 is October, not January. Filter by award type, agency, recipient, keyword, NAICS code, Assistance Listing, PSC, DEF code, or location (place of performance or recipient) to trace spending trends in a specific area; award types default to contracts. Returns per-period totals and optional breakdowns by award category (contracts, grants, direct payments, IDVs, loans, other).

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Exfil surface

Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.

No private-data, untrusted-input, outbound path in this server alone: no tool is tagged outbound.

Reads private data1
Reads files, mail, messages, databases, secrets, calendars or repositories.
Can send data out0
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.

Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.

Remote posture

What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.

Tools listed without sign-in

Looked at Nov 20, 2026, 19:34 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
200
tools/list served without credentials
yes
Tools that change things, listed without credentials
3
Protocol version
2025-06-18
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
not observed
Authorization servers
not observed
Legacy HTTP+SSE transport
no
Dynamic client registration
not observed
Client ID metadata documents
not observed
  • WarnThe whole server

    The endpoint served tools/list without credentials, and the list has 3 tools that change things (3 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://usaspending.caseyjhand.com/mcp: tools/list returned 18 tools to an anonymous client, including usaspending_autocomplete_filters (write), usaspending_spending_by_geography (write), usaspending_spending_by_category (write)

    What this check looks for

Findings

Matches of fixed checks, with the text they matched.

No check matched anything in these tool definitions. That is not a safety guarantee.

History

Changes to its tool definitions and to its remote endpoint, newest first.

  1. Tools9 changed.~9 changedView diff
  2. Tools17 changed.~17 changedView diff
  3. Tools9 changed.~9 changedView diff
  4. Tools9 changed.~9 changedView diff
  5. Tools2 changed.~2 changedView diff

Install

Pinned to version 0.7.3. Values in angle brackets are yours to fill in.

claude mcp add --transport http usaspending https://usaspending.caseyjhand.com/mcp
claude mcp add --transport http usaspending https://usaspending.caseyjhand.com/mcp

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__usaspending__usaspending_list_agencies",
      "mcp__usaspending__usaspending_search_awards",
      "mcp__usaspending__usaspending_get_award",
      "mcp__usaspending__usaspending_get_award_transactions",
      "mcp__usaspending__usaspending_get_award_subawards",
      "mcp__usaspending__usaspending_get_award_federal_accounts",
      "mcp__usaspending__usaspending_get_idv_awards",
      "mcp__usaspending__usaspending_search_recipients",
      "mcp__usaspending__usaspending_get_recipient",
      "mcp__usaspending__usaspending_get_agency",
      "mcp__usaspending__usaspending_spending_over_time",
      "mcp__usaspending__usaspending_disaster_spending",
      "mcp__usaspending__usaspending_get_federal_account",
      "mcp__usaspending__usaspending_get_federal_account_breakdown",
      "mcp__usaspending__usaspending_search_federal_accounts"
    ],
    "ask": [
      "mcp__usaspending__usaspending_autocomplete_filters",
      "mcp__usaspending__usaspending_spending_by_geography",
      "mcp__usaspending__usaspending_spending_by_category"
    ],
    "deny": []
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as usaspending. Review it before you use it.

Capability tags behind 8 rules
  • mcp__usaspending__usaspending_autocomplete_filters untrusted-input
  • mcp__usaspending__usaspending_search_awards untrusted-input
  • mcp__usaspending__usaspending_search_recipients untrusted-input
  • mcp__usaspending__usaspending_spending_by_geography untrusted-input
  • mcp__usaspending__usaspending_spending_by_category untrusted-input
  • mcp__usaspending__usaspending_spending_over_time untrusted-input
  • mcp__usaspending__usaspending_disaster_spending untrusted-input
  • mcp__usaspending__usaspending_search_federal_accounts private-data, untrusted-input

Pin

Keep the decision: fail CI when the tools change, follow every change, or show a badge.

Pin in CI

Fail a build when the tools you reviewed are not the tools the server serves today.

curl -s "https://radar.anlyon.com/api/v1/servers/io.github.cyanheads/usaspending-mcp-server/verify?hash=9f971922b320b52c9ddd5c0b20aa72dc0b42a19565738059c78696eb6e37dc30" | jq -e '.changed == false'

Pinned hash 9f971922b320, from Nov 14, 2026, 07:18 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/io.github.cyanheads/usaspending-mcp-server/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for io.github.cyanheads/usaspending-mcp-server
[![Anlyon Radar](https://radar.anlyon.com/badge/io.github.cyanheads/usaspending-mcp-server.svg)](https://radar.anlyon.com/servers/io.github.cyanheads/usaspending-mcp-server)

Source: MCP Registry · repository. Synced Nov 14, 2026, 07:17 UTC.

usaspending-mcp-server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the usaspending-mcp-server safe?

No scan can promise that. What its definitions show: 18 tools, 3 of which can change, send or delete something, and 0 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.

Can the usaspending-mcp-server delete data?

No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.

Can the usaspending-mcp-server do things that cannot be undone?

No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.

What tools does the usaspending-mcp-server have?

18 tools: 3 write, 15 read. usaspending_autocomplete_filters, usaspending_spending_by_geography, usaspending_spending_by_category, usaspending_list_agencies, usaspending_search_awards, usaspending_get_award, usaspending_get_award_transactions, usaspending_get_award_subawards and 10 more.

Does the usaspending-mcp-server need sign-in or OAuth?

Its remote endpoint listed its tools to an anonymous client, without sign-in.

Does the usaspending-mcp-server send data to outside services?

None of its tools sends, posts or publishes, or takes a recipient, channel or URL.

Has the usaspending-mcp-server changed its tools?

Yes, 5 times since Anlyon Radar started tracking it, most recently on Nov 14, 2026. Every change is on this page and in its Atom feed.