# Govparse Government Data Gateway (remote · api.govparse.io)

Pay-per-call US government data: carrier safety, visa sponsors, contracts, employer risk.

- Trust score: 67/100 (medium)
- Change this week: +5
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-08-03

## Components

- remote · `api.govparse.io`: 67/100 (this document), [markdown](https://verifymcp.io/servers/io-govparse-gateway/api.md), [page](https://verifymcp.io/servers/io-govparse-gateway/api)

## Channel facts

- Endpoint: `https://api.govparse.io/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `1.0.0`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-08-03.

- **Endpoint Security**: 66/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - Authorisation not fully verified: no authorisation is required to call this server, and 78 tool(s) never declared a destructiveHint. The MCP spec treats an absent hint as destructive by default, so we cannot call this surface safe.
  - HTTPS is enforced; there's no plaintext access path.
  - The HSTS (Strict-Transport-Security) header is present.
  - DNSSEC is configured correctly; the domain's records validate against the full chain to the root.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 57/100
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 21417 tokens (~274/item across 78 items; 78 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 27/100
  - Stability observed for 8 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 100/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 100% of tool parameters carry a description.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### Claude

```bash
claude mcp add --transport http io-govparse-gateway https://api.govparse.io/mcp
```

### Codex

```toml
[mcp_servers.io-govparse-gateway]
url = "https://api.govparse.io/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "io-govparse-gateway": {
      "type": "remote",
      "url": "https://api.govparse.io/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add io-govparse-gateway --url https://api.govparse.io/mcp --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  io-govparse-gateway:
    url: "https://api.govparse.io/mcp"
```

### Other

```json
{
  "mcpServers": {
    "io-govparse-gateway": {
      "type": "http",
      "url": "https://api.govparse.io/mcp"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-08-03 (score 67, +1)

No change was recorded against any check on this day. Stability & Change Management went from 23 to 27. That category is still filling its 30-day observation window: 7 days of observed history at the previous scan, 8 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-08-01 (score 66, 0)

- [functional] Schema quality: excellent → good
- [cosmetic] “get_leads” reworded the description of “query”
- [cosmetic] “get_leads” reworded the description of “industry”
- [cosmetic] “build_parse” reworded the description of “query”
- [cosmetic] “build_parse” reworded the description of “industry”

### 2026-07-31 (score 66, +2)

- [functional] We updated how we score, so this day's move reflects our rubric, not a change to the server

### 2026-07-30 (score 64, +1)

- [functional] We updated how we score, so this day's move reflects our rubric, not a change to the server

### 2026-07-29 (score 63, 0)

- [functional regression] Schema quality: 18022 → 21371
- [functional] New tool “nlrb_feed_petitions”
- [functional] New tool “sec_feed_insiders”
- [functional] New tool “trials_feed_starts”
- [functional] New tool “uspto_feed_grants”
- [functional] New tool “va_feed_sdvosb”
- [functional] New tool “sba_feed_loans”
- [functional] New tool “epa_feed_rcra”
- [functional] New tool “fda_feed_approvals”
- [functional] New tool “fda_feed_clearances”
- [functional] New tool “fda_feed_warningletters”
- [functional] New tool “govcon_feed_awards”
- [functional] New tool “govcon_feed_subawards”

### 2026-07-28 (score 63, +1)

- [functional regression] Schema quality: 15752 → 18022
- [functional] Schema quality: good → excellent

### 2026-07-27 (score 62, +2)

- [functional] We updated how we score, so this day's move reflects our rubric, not a change to the server

### 2026-07-26 (score 60)

First indexed and scored.

## MCP tools (78)

### `gateway_health` (~52 tokens)

Is the Govparse Agent Gateway up, and which product datasets does it front? Returns a signed _meta attestation proving the payment-to-attestation pipeline end to end. [price: $0.001/call]

### `fmcsa_carrier_lookup` (~109 tokens)

What does the FMCSA census record show for this USDOT number? Legal/DBA name, entity type, MC number, address, phone, fleet size (power units, drivers), operating status, safety rating, and MCS-150 details for one motor carrier — the registration-of-record lookup behind 'who is this carrier?' [price: $0.01/call]

Input parameters:

- `usdot` (string, required): USDOT number (path parameter). Example: 80321.

### `fmcsa_carrier_search` (~326 tokens)

Which FMCSA-registered carriers match a name, state, entity type, fleet-size range, operating status, or authority status? Trigram name search over 4.4M+ census records — find a carrier's USDOT number, shortlist carriers in a lane state, or screen brokers vs carriers. Up to 100 rows per call, largest fleets first; total_matched caps at 1000 (total_matched_capped=true means narrow your filters). [price: $0.05/row]

Input parameters:

- `authority_status` (string): Only carriers with an operating-authority record in this status: Active | Pending | Inactive | Withdrawn.
- `entity_type` (string): carrier | broker | forwarder (CSV for multiple).
- `limit` (integer): Max rows to return (default 25, cap 100).
- `name` (string): Carrier legal or DBA name fragment, case-insensitive (trigram match). Example: 'knight transportation'.
- `offset` (integer): Rows to skip, for paging.
- `operating_status` (string): A (active) | I (inactive) | P (pending), CSV for multiple.
- `power_units_max` (integer): Maximum fleet size in power units.
- `power_units_min` (integer): Minimum fleet size in power units.
- `sort` (string): field:direction — power_units | total_drivers | usdot. Example: 'power_units:desc'.
- `state` (string): Physical-address state code(s), CSV. Example: 'TX,OK'.

### `fmcsa_carrier_score` (~114 tokens)

What SaferGrade score do FMCSA safety records produce for this carrier, and which components drive it? Returns the 0-100 score with letter grade and confidence, plus the observed inputs: 24-month inspection and out-of-service totals, driver/vehicle OOS rates vs the national baseline, crash summary, and the FMCSA safety rating on file. [price: $0.05/call]

Input parameters:

- `usdot` (string, required): USDOT number (path parameter). Example: 54283.

### `fmcsa_carrier_vet` (~110 tokens)

Is this carrier's paperwork in order before I tender a load? One-call carrier vetting screen: FMCSA census record, operating-authority records and revocations, active insurance filings on file, SaferGrade score, and crash/out-of-service history vs the national baseline — everything a broker checks before onboarding, in a single response. [price: $0.05/call]

Input parameters:

- `usdot` (string, required): USDOT number (path parameter). Example: 54283.

### `fmcsa_carrier360` (~135 tokens)

What does the complete FMCSA file show for this carrier, in one call? The carrier360 dossier by USDOT number: census identity, operating-authority records with status-change history and revocations, active + historical insurance filings, the safety rating and review on file, the SaferGrade score with its component breakdown, 24-month inspection/violation and crash summaries vs national OOS baselines, and a merged recent-events timeline of authority and insurance changes. [price: $0.25/call]

Input parameters:

- `usdot` (string, required): USDOT number (path parameter). Example: 43838.

### `fmcsa_stats_carriers` (~76 tokens)

How big is the FMCSA carrier dataset right now? Headline aggregates: total and active registered carriers, how many have a SaferGrade score and the grade distribution, carriers with inspection/crash history, and the current national driver/vehicle out-of-service baseline rates. [price: $0.001/call]

### `oflc_sponsor_search` (~287 tokens)

Which employers sponsor work visas or green cards? Search DOL OFLC disclosure data by employer name, state, program (LCA/H-1B vs PERM), case status, fiscal year, or SOC occupation code. Returns per-employer filing counts (total, LCA, PERM, certified, denied) — the shortlist behind 'who sponsors H-1B for this role?' [price: $0.05/row]

Input parameters:

- `case_status` (string): CERTIFIED | DENIED | WITHDRAWN | CERTIFIED - WITHDRAWN, CSV for multiple.
- `fiscal_year_max` (integer): Latest federal fiscal year to include.
- `fiscal_year_min` (integer): Earliest federal fiscal year to include.
- `limit` (integer): Max rows (default 25, cap 100).
- `name` (string): Employer name fragment, punctuation-insensitive. Example: 'amazon'.
- `offset` (integer): Rows to skip, for paging.
- `program` (string): LCA (H-1B and related) | PERM (green-card labor certification). CSV for both.
- `soc_code` (string): Only sponsors with LCA filings in this SOC occupation. Example: 15-1252 (software developers).
- `state` (string): Employer HQ state code(s), CSV. Example: 'WA,CA'.

### `oflc_sponsor_dossier` (~142 tokens)

What is this employer's full visa-sponsorship track record? One-call dossier from DOL OFLC filings: LCA and PERM case counts by fiscal year and status, certification vs denial breakdown, annualized offered-wage percentiles (p10-p90), top sponsored SOC occupations, and top worksite states. Accepts a slug (amazon-com-services-llc) or a raw employer name. [price: $0.1/call]

Input parameters:

- `sponsor` (string, required): Employer slug or name (path parameter). Punctuation-insensitive; fuzzy-matched if no exact hit. Example: 'amazon-com-services-llc'.

### `oflc_wages_prevailing` (~134 tokens)

What do employers actually offer for this occupation in this state? Annualized wage percentiles (p10-p90) computed from LCA disclosure filings for an SOC code, optionally narrowed to a worksite state: offered wages, DOL prevailing wages, wage-level distribution, and sample size — the market-rate check behind salary negotiation and offer benchmarking. [price: $0.01/call]

Input parameters:

- `soc` (string, required): SOC occupation code, required. Example: 15-1252 (software developers).
- `state` (string): Worksite state code to narrow the sample. Example: 'CA'.

### `oflc_stats_sponsors` (~64 tokens)

How much visa-sponsorship data is on file? Headline aggregates over DOL OFLC disclosure data: distinct sponsoring employers, LCA and PERM case totals with certified counts, and the fiscal years covered. [price: $0.001/call]

### `govcon_opportunity_search` (~335 tokens)

What federal contract solicitations and grant opportunities can my business bid on right now? Unified search over SAM.gov contract opportunities and Grants.gov postings: keyword, awarding agency, NAICS prefix, set-aside (SBA, 8A, WOSB, SDVOSBC...), place-of-performance state, and posted/deadline date windows. One call answers 'find open federal money that fits us'. [price: $0.05/row]

Input parameters:

- `agency` (string): Awarding department/agency name fragment. Example: 'defense'.
- `deadline_after` (string): Response/close deadline on or after this date (YYYY-MM-DD).
- `deadline_before` (string): Response/close deadline on or before this date (YYYY-MM-DD).
- `kind` (string): contract | grant (CSV for both; default both).
- `limit` (integer): Max rows (default 25, cap 100).
- `naics` (string): NAICS code prefix(es), CSV — contracts only. Example: '5415' covers all IT services.
- `offset` (integer): Rows to skip, for paging.
- `posted_after` (string): Posted/open date on or after this date (YYYY-MM-DD).
- `q` (string): Keyword in the opportunity title. Example: 'cybersecurity'.
- `setaside` (string): Set-aside code(s), CSV — contracts only: SBA, 8A, WOSB, SDVOSBC, HZC...
- `state` (string): Place-of-performance state code(s), CSV — contracts only.

### `govcon_opportunity_detail` (~113 tokens)

What are the full details of this federal opportunity? Complete record for one SAM.gov contract notice (by notice ID) or Grants.gov opportunity (by opportunity ID or number): solicitation number, type, set-aside, NAICS, place of performance, response deadline, award floor/ceiling, eligibility, description, and the official link. [price: $0.01/call]

Input parameters:

- `noticeId` (string, required): SAM.gov notice ID, or Grants.gov opportunity ID / opportunity number (path parameter).

### `govcon_award_search` (~320 tokens)

Who has the government actually paid, for what, and how much? Search USAspending federal award history (contracts and assistance) by recipient name or UEI, awarding agency, NAICS prefix, CFDA number, amount range, and action-date window — competitor award research, incumbent lookup before a recompete, and grant-winner analysis in one call. [price: $0.05/row]

Input parameters:

- `agency` (string): Awarding agency or sub-agency name fragment. Example: 'defense'.
- `amount_max` (number): Maximum award amount in USD.
- `amount_min` (number): Minimum award amount in USD.
- `awarded_after` (string): Action date on or after (YYYY-MM-DD).
- `awarded_before` (string): Action date on or before (YYYY-MM-DD).
- `cfda` (string): CFDA / Assistance Listing number, exact match (assistance awards).
- `kind` (string): contract | assistance (grants, loans, direct payments). CSV for both.
- `limit` (integer): Max rows (default 25, cap 100).
- `naics` (string): NAICS code prefix(es), CSV.
- `offset` (integer): Rows to skip, for paging.
- `recipient` (string): Award recipient name fragment. Example: 'lockheed'.
- `sort` (string): field:direction — amount | action_date. Default 'amount:desc'.
- `uei` (string): Recipient Unique Entity Identifier (SAM UEI), exact match.

### `govcon_stats_opportunities` (~77 tokens)

How much open federal money is in the dataset right now? Headline aggregates: active SAM.gov contract opportunities (and how many still have an open deadline), Grants.gov opportunities by status (posted vs forecasted), and USAspending award records on file with their total dollar amount. [price: $0.001/call]

### `firmstanding_business360_lookup` (~171 tokens)

What do federal records show about this company, across every dataset at once? Flagship employer dossier by name + state: canonical identity from the entity graph, plus per-source sections — OFLC visa filings, WHD wage enforcement (back wages), OSHA inspections/violations (penalties), EPA ECHO facilities, FMCSA carriers, LEIE/SAM exclusion hits — match tiers and confidence disclosed. Public records, name/address-matched; NOT a consumer report, no FCRA use. [price: $0.25/call]

Input parameters:

- `name` (string, required): Company name — legal suffixes and punctuation are normalized away ('Tyson Foods, Inc.' matches 'TYSON FOODS').
- `state` (string, required): Two-letter state code the entity is keyed to. Example: 'AR'.

### `firmstanding_business360` (~148 tokens)

What do federal records show about this exact employer entity? The business360 dossier by entity UUID (from /v1/companies/search or /v1/business360/lookup): canonical identity plus per-source sections — OFLC visa filings, WHD wage enforcement, OSHA inspections/violations, EPA ECHO compliance, FMCSA carriers, LEIE/SAM exclusions — with match tiers, confidence, and per-source vintages disclosed. Public records; NOT a consumer report, no FCRA use. [price: $0.25/call]

Input parameters:

- `entityId` (string, required): Employer entity UUID (path parameter), as returned by /v1/companies/search or a business360 lookup.

### `firmstanding_screen_exclusions` (~167 tokens)

Is this company on a federal exclusion or debarment list? One-call screen of HHS OIG LEIE and SAM.gov exclusions by normalized business name (state optional): attested records_found verdict, per-list hit and currently-active counts, and the matching records with exclusion type, agency, and dates — plus entity-linked hits with disclosed match confidence. Name-matched public records; NOT a consumer report, no FCRA-regulated use (employment/credit/tenancy). [price: $0.1/call]

Input parameters:

- `name` (string, required): Business name to screen — legal suffixes and punctuation normalized ('Purdue Pharma, L.P.' matches 'PURDUE PHARMA').
- `state` (string): Optional two-letter state code to narrow the screen.

### `firmstanding_screen_enforcement` (~158 tokens)

Does this employer have federal enforcement history on file? One-call WHD + OSHA + EPA screen by company name and state: attested records_found verdict with per-source counts, severity totals (WHD back wages and penalties, OSHA current violation penalties, EPA ECHO 5-year penalties and SNC/HPV flags), most-recent action dates, and the latest records — with match tiers and confidence disclosed. Public records, name/address-matched; NOT a consumer report, no FCRA use. [price: $0.1/call]

Input parameters:

- `name` (string, required): Company name — legal suffixes and punctuation normalized.
- `state` (string, required): Two-letter state code the employer is keyed to. Example: 'AR'.

### `firmstanding_company_search` (~348 tokens)

Which employers appear in federal enforcement and compliance records? Search 3.7M resolved employer entities by name (suffix/punctuation-insensitive, trigram-indexed), state, linked source (whd, osha, echo, oflc, fmcsa, leie, sam_exclusions), NAICS prefix, or source-count range. Returns entity IDs for the business360 dossier plus identity, address, and source/record counts. total_matched caps at 1000 — narrow when capped. Not a consumer report; no FCRA use. [price: $0.05/row]

Input parameters:

- `limit` (integer): Max rows (default 25, cap 100).
- `naics` (string): NAICS code prefix(es), CSV. Example: '3116' covers animal processing.
- `name` (string): Company name fragment — normalized before matching ('Swift Transportation Co. of Arizona, LLC' → 'SWIFT TRANSPORTATION'). Example: 'swift transportation'.
- `offset` (integer): Rows to skip, for paging.
- `sort` (string): field:direction — source_count | record_count | last_seen. Default 'source_count:desc'.
- `source` (string): Only entities with linked records in these sources (CSV): whd | osha | echo | oflc | fmcsa | leie | sam_exclusions.
- `source_count_max` (integer): Maximum number of distinct federal sources the entity appears in.
- `source_count_min` (integer): Minimum number of distinct federal sources the entity appears in (1-7).
- `state` (string): Entity state code(s), CSV. Example: 'AZ'.

### `firmstanding_stats_companies` (~90 tokens)

How big is the FirmStanding employer entity graph right now? Headline aggregates: resolved employer entities and how many appear in 2+ federal sources, states covered, record links by source system (whd, osha, echo, oflc, fmcsa, leie, sam_exclusions), and raw records on file per dataset. [price: $0.001/call]

### `partd_feed_prescribing` (~380 tokens)

Which prescribers just entered a drug class this Medicare Part D year — e.g. new GLP-1 prescribers in PY2024? Bulk feed over CMS Part D Prescribers by Provider and Drug, aggregated per NPI to a drug class and differenced PY2023 → PY2024: direction=new_entrant returns new-to-class prescribers, de_adopter those whose class volume fell. Flat rows, cursor-paginated up to 1000/page, by drug_class, specialty, and state. Observational public Part D records — never a clinical recommendation. [price: $0.05/row]

Input parameters:

- `cursor` (string): Opaque page cursor — pass the previous page's next_cursor unchanged.
- `direction` (string): new_entrant (default; wrote the class in PY2024 with ≤ max_prior_claims in PY2023 — the new-to-class buying moment) | de_adopter (class volume fell ≥ min_pct_drop%).
- `drug_class` (string): Therapeutic drug class: GLP-1 | SGLT2 | DPP-4 | Statin. Default GLP-1.
- `limit` (integer): Rows per page (default 500, cap 1000).
- `max_prior_claims` (integer): new_entrant only: maximum PY2023 class claims allowed (0 = truly new to the class).
- `min_claims_2024` (integer): Minimum class claims in PY2024.
- `min_pct_drop` (integer): de_adopter only: minimum percent decline in class volume PY2023 → PY2024.
- `specialty` (string): Prescriber specialty (Prscrbr_Type), e.g. Endocrinology.
- `state` (string): Prescriber state code (two letters).

### `partd_prescriber_search` (~212 tokens)

Who is this prescriber, and what do they prescribe most in Medicare Part D? Look up a prescriber by NAME (or exact NPI): NPPES identity — specialty, location — plus that NPI's latest-year Part D summary (top drug, total claims, distinct drugs, and which tracked classes GLP-1/SGLT2/DPP-4/Statin they write). Optionally narrow by state and specialty. Observational public CMS/NPPES records — never a clinical recommendation. Cross-gateway via the Healthparse Part D API. [price: $0.05/row]

Input parameters:

- `name` (string): Prescriber name — last-name prefix (e.g. Smith) or first+last (e.g. John Smith).
- `npi` (string): Exact 10-digit National Provider Identifier.
- `specialty` (string): NPPES taxonomy / specialty substring.
- `state` (string): Prescriber state code(s), two letters, comma-separated.

### `firmstanding_feed_enforcement` (~255 tokens)

Which employers were hit with new federal enforcement actions since a given date? Bulk feed of OSHA inspections (violated standards, penalties, abatement dates), EPA ECHO formal enforcement actions (SNC/HPV flags, penalties), and DOL WHD concluded wage cases (back wages, statutes) — flat rows, cursor-paginated up to 1000/page, each resolved to the FirmStanding entity graph where a match exists, with tier and confidence disclosed. Public records; NOT a consumer report, no FCRA use. [price: $0.05/row]

Input parameters:

- `agency` (string): Restrict to source agencies (CSV): osha | epa | whd. Default all three.
- `cursor` (string): Opaque page cursor — pass the previous page's next_cursor unchanged.
- `limit` (integer): Rows per page (default 500, cap 1000).
- `naics` (string): NAICS code prefix(es), CSV. Example: '3116' covers animal processing.
- `since` (string, required): Include events dated on or after this date (YYYY-MM-DD). Required.
- `state` (string): Facility/employer state code(s), CSV.

### `oflc_feed_workforce` (~358 tokens)

Which employers filed U.S. work-visa cases — H-1B LCA, PERM, and the seasonal-labor programs H-2A, H-2B, and CW-1 — for what roles, wages, worksites, and how many seasonal workers over what period of need? Employer-level bulk feed over DOL OFLC disclosure files: one flat row per case decided since a date, cursor-paginated up to 1000/page, filterable by program, worksite state, SOC, or employer name. Rows add employer FEIN where filed, SEC CIK/ticker where public, and as_published_date. [price: $0.05/row]

Input parameters:

- `cursor` (string): Opaque page cursor — pass the previous page's next_cursor unchanged.
- `employer` (string): Employer name, punctuation-insensitive exact match ('Amazon.com Services LLC' matches 'AMAZON COM SERVICES LLC').
- `limit` (integer): Rows per page (default 500, cap 1000).
- `program` (string): Which OFLC programs to include (CSV): lca | perm (default; professional workers) or the seasonal-labor programs h2a (agricultural) | h2b (non-ag seasonal) | cw1 (CNMI). Seasonal rows add workers_requ…
- `since` (string, required): Include cases with a decision date on or after this date (YYYY-MM-DD). Required.
- `soc` (string): SOC occupation code. Example: 15-1252 (software developers).
- `state` (string): Worksite state code(s), CSV.

### `fmcsa_feed_carriers` (~225 tokens)

What does the full FMCSA carrier census look like, carrier by carrier, with authority and safety context attached? Bulk analytics feed in USDOT order: legal/DBA name, operating status, authority statuses on file, active insurance filings, safety rating, SaferGrade score, 24-month SMS inspection/out-of-service/crash summaries, fleet size, and address — flat rows, cursor-paginated up to 1000/page, filterable by state, minimum fleet size, or census add date. [price: $0.05/row]

Input parameters:

- `added_since` (string): Only carriers added to the census on or after this date (YYYY-MM-DD) — the census carries no per-record modified date.
- `cursor` (string): Opaque page cursor — pass the previous page's next_cursor unchanged.
- `limit` (integer): Rows per page (default 500, cap 1000).
- `min_power_units` (integer): Minimum fleet size (power units).
- `state` (string): Physical-address state code(s), CSV.

### `firmstanding_feed_distress` (~334 tokens)

Which businesses carry public-record distress markers? One observational row per subject across ten tracked marker types — WARN notice, Form 5500 participant drop, final filing, or Schedule H asset decline, IRS auto-revocation, SEC NT / 8-K / going-concern filings, FMCSA authority revocation, active enforcement case — with per-marker counts, dates, and details. Ticker/CIK-mapped where identifier-verified, basis disclosed. Counts as filed, never a score. NOT a consumer report, no FCRA use. [price: $0.05/row]

Input parameters:

- `cursor` (string): Opaque page cursor — pass the previous page's next_cursor unchanged.
- `limit` (integer): Rows per page (default 500, cap 1000).
- `marker` (string): Only subjects carrying this marker key: warn_notice_180d | active_participants_drop_20pct_yoy | final_form5500_filing | irs_auto_revocation_1y | nt_late_filing_180d | form8k_distress_item_180d | goin…
- `min_markers` (integer): Only subjects carrying at least this many distinct marker types (1-9). Default 1.
- `since` (string): Only subjects with a marker event dated on or after this date (YYYY-MM-DD).
- `state` (string): Subject state code(s), CSV. Some subjects carry no state (identified by EIN/CIK only).

### `sec_feed_insiders` (~283 tokens)

Which public companies are insiders buying as a cluster right now? Bulk feed over SEC Form 4 open-market BUY clusters — issuers where several distinct officers/directors/10%-owners bought in a trailing 30- or 90-day window. Flat rows newest-first by the cluster's most recent buy, cursor-paginated up to 1000/page, filterable by ticker, exchange, min insiders, min value, window, or since. Every issuer is a public company with ticker and CIK. Observational filings as filed, never advice. [price: $0.05/row]

Input parameters:

- `cursor` (string): Opaque page cursor — pass the previous page's next_cursor unchanged.
- `exchange` (string): Listing exchange(s), CSV: NYSE, Nasdaq, OTC...
- `limit` (integer): Rows per page (default 500, cap 1000).
- `min_insiders` (integer): Minimum distinct insiders in the cluster (>= 2).
- `min_value` (integer): Minimum total $ purchased across the cluster window.
- `since` (string): Only clusters whose most recent buy is on or after this date (YYYY-MM-DD).
- `ticker` (string): Issuer ticker(s), CSV.
- `window` (string): Trailing cluster window in days: 30 | 90. Default 90.

### `trials_feed_starts` (~327 tokens)

Which companies just opened a new clinical trial or site? Bulk feed over NIH ClinicalTrials.gov interventional, industry-sponsored studies, newest-first by registry-posting date (the R&D / commercial-growth signal). Cursor-paginated up to 1000/page, filterable by sponsor, phase, status, condition, intervention type, site state, or since. Rows carry sponsor, phase, status, conditions, enrollment, dates, site count, and the US states sited. Observational registry records, never a prediction. [price: $0.05/row]

Input parameters:

- `condition` (string): Indication / condition fragment (matched against the trial's conditions).
- `cursor` (string): Opaque page cursor — pass the previous page's next_cursor unchanged.
- `intervention_type` (string): Intervention type(s), CSV: DRUG | DEVICE | BIOLOGICAL | PROCEDURE...
- `limit` (integer): Rows per page (default 500, cap 1000).
- `phase` (string): Trial phase(s), CSV: PHASE1 | PHASE2 | PHASE3 | PHASE4 | EARLY_PHASE1 | NA.
- `since` (string): Only trials first posted to the registry on or after this date (YYYY-MM-DD).
- `sponsor` (string): Lead sponsor company name fragment.
- `state` (string): Has a trial site in these US state code(s), CSV.
- `status` (string): Overall status, CSV: RECRUITING | NOT_YET_RECRUITING | COMPLETED | TERMINATED...

### `govcon_feed_awards` (~263 tokens)

Which companies just won federal money? Bulk feed over USAspending prime awards (contracts + assistance), newest-first by action date — a funded-growth signal for the awardee. Flat rows, cursor-paginated up to 1000/page, filterable by NAICS, awarding agency, minimum amount, award kind, or since. Each row carries the awardee, agency, award amount (with a banded size bucket), NAICS, dates, and UEI. Observational public records as filed. [price: $0.05/row]

Input parameters:

- `agency` (string): Awarding agency or sub-agency name fragment.
- `cursor` (string): Opaque page cursor — pass the previous page's next_cursor unchanged.
- `kind` (string): contract | assistance (grants, loans, direct payments). CSV for both; default both.
- `limit` (integer): Rows per page (default 500, cap 1000).
- `min_amount` (integer): Minimum award amount in USD.
- `naics` (string): NAICS code prefix(es), CSV. Example: '5415' covers IT services.
- `since` (string): Only awards with an action date on or after this date (YYYY-MM-DD).

### `uspto_feed_grants` (~288 tokens)

Which companies are granting a cluster of patents? Bulk feed over USPTO / PatentsView granted-patent assignees, ordered by grant count in the loaded window — an assignee with a cluster of grants is an R&D/growth signal. Cursor-paginated up to 1000/page, filterable by state, minimum grants, CPC subclass, first-time-assignee only, or since. Rows carry the assignee, grant count, CPC tech areas, grant window, inventor count, and a first_time_assignee flag. CC BY 4.0 attribution (in _meta.notice). [price: $0.05/row]

Input parameters:

- `cpc` (string): CPC subclass(es), CSV — the technology area. Example: G06N (AI/ML).
- `cursor` (string): Opaque page cursor — pass the previous page's next_cursor unchanged.
- `limit` (integer): Rows per page (default 500, cap 1000).
- `min_grants` (integer): Minimum grant count in the window.
- `new_only` (string): true = only first-time (newly-observed) assignees granting in this window.
- `since` (string): Only assignees whose latest grant is on or after this date (YYYY-MM-DD).
- `state` (string): Assignee state code(s), CSV.

### `sba_feed_loans` (~250 tokens)

Which small businesses were just approved for an SBA loan? Bulk feed over SBA 7(a) & 504 FOIA loan approvals, newest-first by approval date — a business approved for a federal loan is funded to grow. Cursor-paginated up to 1000/page, filterable by borrower state, program (7A/504), NAICS, minimum amount, or since. Rows carry the borrower, city/state, program, approval amount (banded), lender/CDC, NAICS, and jobs supported. Observational public records as disclosed. [price: $0.05/row]

Input parameters:

- `cursor` (string): Opaque page cursor — pass the previous page's next_cursor unchanged.
- `limit` (integer): Rows per page (default 500, cap 1000).
- `min_amount` (integer): Minimum gross approval amount in USD.
- `naics` (string): NAICS code prefix(es), CSV.
- `program` (string): Loan program: 7A | 504.
- `since` (string): Only loans approved on or after this date (YYYY-MM-DD).
- `state` (string): Borrower state code(s), CSV.

### `fda_feed_approvals` (~237 tokens)

Which drugs did the FDA just approve, and who is the sponsor? Bulk feed over openFDA Drugs@FDA approvals, newest-first by approval date — a new approval or supplement is a commercial-launch moment. Cursor-paginated up to 1000/page, filterable by sponsor, submission type (ORIG/SUPPL), marketing status, or since. Rows carry the sponsor, application number/type, brand names, active ingredients, dosage/route, and marketing status. Observational public records, never a clinical recommendation. [price: $0.05/row]

Input parameters:

- `cursor` (string): Opaque page cursor — pass the previous page's next_cursor unchanged.
- `limit` (integer): Rows per page (default 500, cap 1000).
- `marketing_status` (string): Marketing status fragment (Prescription, Over-the-counter, Discontinued...).
- `since` (string): Only approvals dated on or after this date (YYYY-MM-DD).
- `sponsor` (string): Sponsor company name fragment.
- `submission_type` (string): ORIG (original approval) | SUPPL (supplement).

### `fda_feed_clearances` (~267 tokens)

Which medical devices did the FDA just clear or approve, and who is the applicant? Bulk feed over openFDA 510(k) clearances and PMA approvals, newest-first by decision date — a new clearance is a device launch. Cursor-paginated up to 1000/page, filterable by applicant, state, product code, pathway (510k/pma), advisory committee, or since. Rows carry the applicant, device name, product code, advisory committee, pathway, and decision. Observational public records, never a recommendation. [price: $0.05/row]

Input parameters:

- `applicant` (string): Applicant (manufacturer) name fragment.
- `committee` (string): FDA advisory-committee / specialty panel fragment (Cardiovascular, Radiology...).
- `cursor` (string): Opaque page cursor — pass the previous page's next_cursor unchanged.
- `limit` (integer): Rows per page (default 500, cap 1000).
- `pathway` (string): Regulatory pathway: 510k | pma.
- `product_code` (string): FDA product code (3-letter).
- `since` (string): Only decisions dated on or after this date (YYYY-MM-DD).
- `state` (string): Applicant state code(s), CSV.

### `fda_feed_warningletters` (~298 tokens)

Which firms just received an FDA Warning Letter — an open compliance-gap and remediation clock? Bulk feed over the FDA Warning Letters index, newest-first by issue date: recipient firm, product area (drugs/devices/food/tobacco…), issuing office, alleged-violation subject, and the response / close-out clock. Flat rows, cursor-paginated up to 1000/page, filterable by product_area, subject, close-out state, or since. Observational records — an alleged violation, never an adjudication. [price: $0.05/row]

Input parameters:

- `closed` (string): true = closed-out letters only; false = open (no close-out) only.
- `cursor` (string): Opaque page cursor — pass the previous page's next_cursor unchanged.
- `limit` (integer): Rows per page (default 500, cap 1000).
- `office` (string): Issuing FDA center / district office fragment (e.g. Center for Drug Evaluation and Research).
- `product_area` (string): Product area(s), CSV: drugs | devices | biologics | food | dietary-supplement | cosmetics | tobacco | veterinary | other.
- `since` (string): Only letters issued on or after this date (YYYY-MM-DD).
- `subject` (string): Alleged-violation subject fragment (e.g. CGMP, Adulterated, Unapproved New Drug).

### `govcon_feed_subawards` (~294 tokens)

Which companies just received a federal subaward — and which are getting their FIRST one? Bulk feed over USAspending subawards (the File F prime→sub cascade), newest-first by action date: sub-recipient, prime, agency, amount (banded), NAICS, and date. new_only isolates first-time subawardees (no prime award on record) — the new-to-govcon growth signal. Flat rows, cursor-paginated up to 1000/page, filterable by state, NAICS, agency, minimum amount, or since. Observational public records as filed. [price: $0.05/row]

Input parameters:

- `agency` (string): Awarding agency or sub-agency name fragment.
- `cursor` (string): Opaque page cursor — pass the previous page's next_cursor unchanged.
- `limit` (integer): Rows per page (default 500, cap 1000).
- `min_amount` (integer): Minimum subaward amount in USD.
- `naics` (string): NAICS code prefix(es), CSV.
- `new_only` (string): true = only first-time subawardees (sub-recipients with NO prime award on record).
- `since` (string): Only subawards with an action date on or after this date (YYYY-MM-DD).
- `state` (string): Place-of-performance state code(s), CSV.

### `va_feed_sdvosb` (~266 tokens)

Which veteran-owned (SDVOSB / VOSB) vendors just showed fresh federal traction? Bulk feed over the verified veteran-owned vendor roster, newest-first by most recent award date: vendor, state, NAICS, certification, and its VA-specific + total federal award totals (banded). A set-aside-eligible vendor with recent traction is a teaming / growth target. Flat rows, cursor-paginated up to 1000/page, filterable by state, NAICS, certification, VA award, or since. Observational public records as filed. [price: $0.05/row]

Input parameters:

- `certification` (string): sdvosb | vosb.
- `cursor` (string): Opaque page cursor — pass the previous page's next_cursor unchanged.
- `limit` (integer): Rows per page (default 500, cap 1000).
- `min_va_award` (integer): Minimum total $ this vendor has won from the VA.
- `naics` (string): NAICS code prefix(es), CSV — matches the vendor's award NAICS.
- `since` (string): Only vendors whose most recent award is on or after this date (YYYY-MM-DD).
- `state` (string): Vendor state code(s), CSV.

### `epa_feed_rcra` (~308 tokens)

What does the full RCRAInfo hazardous-waste handler roster look like, handler by handler? Bulk feed over EPA RCRAInfo in handler-id order: facility, city, state, generator status, roles (LQG/SQG/VSQG/transporter/TSDF), and NAICS — a large-quantity generator or TSDF is an EHS-services / environmental-consulting target. Flat rows, cursor-paginated up to 1000/page, filterable by state, generator status, NAICS, or LQG/TSDF/transporter flags. A standing roster (no per-record registration date). [price: $0.05/row]

Input parameters:

- `active` (string): false = include inactive handler sites (default: active only).
- `cursor` (string): Opaque page cursor — pass the previous page's next_cursor unchanged.
- `generator_status` (string): Generator status(es), CSV: LQG | SQG | VSQG | Non-Generator.
- `limit` (integer): Rows per page (default 500, cap 1000).
- `lqg` (string): true = large-quantity generators only.
- `naics` (string): NAICS code prefix(es), CSV.
- `state` (string): State code(s), CSV.
- `transporter` (string): true = hazardous-waste transporters only.
- `tsdf` (string): true = treatment/storage/disposal facilities only.

### `nlrb_feed_petitions` (~268 tokens)

Which employers just drew a union representation petition? Bulk feed over NLRB cases, newest-first by filing date, defaulting to representation petitions (R cases): employer, case type, region, filing date, and unit size — a labor-relations moment for PEOs, staffing, and labor counsel. category widens to ULP charges or all cases. Flat rows, cursor-paginated up to 1000/page, filterable by state, region, or union. Observational records AS FILED — a petition is a filing, never an adjudication. [price: $0.05/row]

Input parameters:

- `category` (string): representation (default; R-case petitions) | ulp (unfair-labor-practice charges) | unit | all.
- `cursor` (string): Opaque page cursor — pass the previous page's next_cursor unchanged.
- `limit` (integer): Rows per page (default 500, cap 1000).
- `region` (string): NLRB region code(s), CSV (two-digit).
- `since` (string): Only cases filed on or after this date (YYYY-MM-DD).
- `state` (string): Employer state code(s), CSV.
- `union` (string): Union / petitioner name fragment.

### `fda_approvals_drugs` (~316 tokens)

Which companies just won an FDA drug approval or supplement? Search openFDA drug/drugsfda applications by sponsor company, application type (NDA/ANDA/BLA), route, marketing status, most-recent approval date (since), or original market-entry date (first_approval_after). Returns the applicant company (entity-resolved where possible), brand/ingredient/dosage, and approval dates — a day-one commercial/partnership target list for pharma sellers and CROs. openFDA public-domain records. [price: $0.05/row]

Input parameters:

- `application_type` (string): NDA | ANDA | BLA (CSV).
- `company` (string): Applicant/sponsor company — suffix/punctuation-insensitive.
- `entity_id` (string): Resolved employer entity UUID (pivots to business360).
- `first_approval_after` (string): Original approval (market entry) on/after this date (YYYY-MM-DD) — surfaces new market entrants.
- `limit` (integer): Max rows (default 25, cap 100).
- `marketing_status` (string): Prescription | Over-the-counter | Discontinued.
- `offset` (integer): Rows to skip.
- `route` (string): Route-of-administration fragment.
- `since` (string): Most recent approval/supplement on/after this date (YYYY-MM-DD).
- `sort` (string): latest_approval_date | original_approval_date :asc|:desc. Default latest_approval_date:desc.

### `fda_clearances_devices` (~284 tokens)

Which companies just cleared or got FDA approval for a medical device? Search openFDA device 510(k) clearances and PMA premarket approvals by applicant company, pathway (510k/pma), FDA product code, advisory-committee specialty (e.g. Cardiovascular), applicant state, or decision date (since). Returns the applicant company (entity-resolved where possible), device name, product code, and decision date — a fresh target list for device and component sellers. openFDA public-domain records. [price: $0.05/row]

Input parameters:

- `advisory_committee` (string): Medical specialty panel fragment.
- `company` (string): Applicant company — suffix/punctuation-insensitive.
- `entity_id` (string): Resolved employer entity UUID.
- `limit` (integer): Max rows (default 25, cap 100).
- `offset` (integer): Rows to skip.
- `pathway` (string): 510k | pma (CSV).
- `product_code` (string): FDA product code(s), CSV.
- `since` (string): Decision (clearance/approval) date on/after this date (YYYY-MM-DD).
- `sort` (string): decision_date | date_received :asc|:desc. Default decision_date:desc.
- `state` (string): Applicant state code(s), CSV (510(k) only).

### `fda_recalls_search` (~283 tokens)

Which companies have an open or recent FDA drug/device recall? Search openFDA drug + device enforcement (recall) reports by recalling firm, product type (Drugs/Devices), classification (Class I/II/III), status, state, or initiation date (since). Returns the recalling company (entity-resolved where possible), classification, reason, product, and dates — a remediation and competitive-displacement trigger for quality and med-device sellers. openFDA public-domain records; observational. [price: $0.05/row]

Input parameters:

- `classification` (string): Class I | Class II | Class III (CSV).
- `company` (string): Recalling firm — suffix/punctuation-insensitive.
- `entity_id` (string): Resolved employer entity UUID.
- `limit` (integer): Max rows (default 25, cap 100).
- `offset` (integer): Rows to skip.
- `product_type` (string): Drugs | Devices (CSV).
- `since` (string): Recall initiated on/after this date (YYYY-MM-DD).
- `sort` (string): recall_initiation_date | report_date :asc|:desc. Default recall_initiation_date:desc.
- `state` (string): Recalling-firm state code(s), CSV.
- `status` (string): Ongoing | Terminated | Completed (CSV).

### `va_facilities_search` (~220 tokens)

Where are the VA's facilities, and of what type? Search Veterans Health Administration sites of care (medical centers, CBOCs, vet centers, health care centers) by state, city, name, or facility type. Returns location, address, coordinates, and — when the VA Lighthouse Facilities API key is configured — classification, health services, and operating status. A territory map for medical-device, imaging, staffing, and construction sellers into the VA. Public VA/HRSA records. [price: $0.05/row]

Input parameters:

- `city` (string): City name fragment.
- `facility_type` (string): Facility type(s), CSV: vet_center, medical_center, outpatient_clinic, health_care_center, primary_care_cboc, multi_specialty_cboc.
- `limit` (integer): Max rows (default 25, cap 100).
- `name` (string): Facility name fragment.
- `offset` (integer): Rows to skip, for paging.
- `state` (string): State code(s), CSV.

### `va_vendors_search` (~276 tokens)

Which verified veteran-owned businesses can I target or team with? Search SDVOSB / VOSB firms (from federal award records carrying the veteran-owned certification flag) by state, name, certification, NAICS, UEI, minimum dollars won FROM the VA, or most-recent award date (since). Returns the vendor, award footprint, VA-won total, and resolved entities-graph id — a seller target list AND a set-aside teaming signal. Public USAspending records; not a certification-status determination. [price: $0.05/row]

Input parameters:

- `certification` (string): sdvosb | vosb (CSV for both).
- `limit` (integer): Max rows (default 25, cap 100).
- `min_va_award` (string): Minimum total $ this vendor has won FROM the VA.
- `naics` (string): NAICS code prefix(es), CSV — matches any of the vendor's award NAICS.
- `name` (string): Vendor name fragment.
- `offset` (integer): Rows to skip, for paging.
- `since` (string): Most recent award on or after this date (YYYY-MM-DD).
- `state` (string): Vendor state code(s), CSV.
- `uei` (string): Recipient Unique Entity Identifier (SAM UEI).

### `va_opportunities_search` (~259 tokens)

What is the Department of Veterans Affairs soliciting right now? A VA-focused slice of the SAM.gov contract-opportunity pipeline (department = Veterans Affairs): search active VA solicitations by keyword, NAICS, set-aside (SDVOSBC, VOSB, SBA...), place-of-performance state, or posted date. Returns the solicitation, NAICS/PSC, set-aside, deadline, and link — VA medical-center construction, equipment, and services buys for device, construction, and staffing sellers. [price: $0.05/row]

Input parameters:

- `limit` (integer): Max rows (default 25, cap 100).
- `naics` (string): NAICS code prefix(es), CSV.
- `offset` (integer): Rows to skip, for paging.
- `posted_after` (string): Posted on or after this date (YYYY-MM-DD).
- `q` (string): Keyword in the solicitation title.
- `setaside` (string): Set-aside code(s), CSV: SDVOSBC, VOSB, SBA, 8A...
- `since` (string): Posted on or after this date (YYYY-MM-DD).
- `state` (string): Place-of-performance state code(s), CSV.

### `sec_insiders_clusters` (~322 tokens)

Where are corporate insiders buying or selling as a cluster? Search SEC Form 4 clusters — issuers where several distinct officers/directors/10%-owners transacted the same direction in a trailing 30- or 90-day window (buy=accumulation, sell=distribution). Filter by ticker, direction, window, minimum insiders, minimum value, or exchange. Returns issuer ticker/company/exchange, insider count, and total shares/value. Public SEC filings as filed — observational, not advice. [price: $0.05/row]

Input parameters:

- `direction` (string): buy (open-market purchase cluster) | sell (open-market sale cluster).
- `exchange` (string): Listing exchange(s), CSV: NYSE, Nasdaq, OTC...
- `issuer_cik` (string): SEC issuer CIK.
- `limit` (integer): Max rows (default 25, cap 100).
- `min_insiders` (string): Minimum distinct insiders in the cluster (>= 2).
- `min_value` (string): Minimum total $ transacted in the window.
- `offset` (integer): Rows to skip, for paging.
- `since` (string): Most recent transaction in the cluster on/after this date (YYYY-MM-DD).
- `sort` (string): insider_count | total_value | txn_count | last_txn_date :asc|:desc. Default insider_count:desc.
- `ticker` (string): Issuer ticker(s), CSV.
- `window` (string): Trailing window in days: 30 | 90.

### `sec_insiders_transactions` (~366 tokens)

What did a company's insiders actually file? Search individual SEC Form 4 transaction lines by ticker, issuer CIK, insider name, direction (buy/sell/other), transaction code (P/S/A/M/F/G), role (officer/director/ten_percent), state, security type, minimum value, or date. Returns the insider, role, security, code, shares, price, and value per filing — the raw feed behind the cluster signal. Public SEC filings as filed — observational, not advice. [price: $0.05/row]

Input parameters:

- `direction` (string): buy | sell | other (grants, option exercises, tax withholding...).
- `issuer_cik` (string): SEC issuer CIK.
- `limit` (integer): Max rows (default 25, cap 100).
- `min_value` (string): Minimum transaction value in $ (shares × price).
- `offset` (integer): Rows to skip, for paging.
- `owner` (string): Reporting-insider name fragment.
- `role` (string): Insider role: officer | director | ten_percent.
- `security_type` (string): nonderiv (common shares) | deriv (options/warrants).
- `since` (string): Transaction on/after this date (YYYY-MM-DD).
- `sort` (string): txn_date | txn_value | shares | filing_date :asc|:desc. Default txn_date:desc.
- `state` (string): Insider state code(s), CSV.
- `ticker` (string): Issuer ticker(s), CSV (as filed on the Form 4).
- `txn_code` (string): SEC transaction code(s), CSV: P (buy), S (sell), A (grant), M, F, G...

### `govcon_subaward_search` (~521 tokens)

When a prime wins federal work, who does it flow subcontracts and subgrants down to? Search USAspending subawards (File F prime → sub cascade) by sub-recipient or prime name/UEI, prime award id, awarding agency, NAICS, PSC, subaward type (contract/grant), place-of-performance state, amount, or action date. Set new_subawardee=true for first-time subawardees — subs with NO prime award on record, coming up through subcontracts (a bonding/staffing/lending/teaming lead). Public federal records. [price: $0.05/row]

Input parameters:

- `agency` (string): Awarding agency or sub-agency name fragment. Example: 'navy'.
- `amount_max` (number): Maximum subaward amount in USD.
- `amount_min` (number): Minimum subaward amount in USD.
- `awarded_after` (string): Subaward action date on or after (YYYY-MM-DD).
- `awarded_before` (string): Subaward action date on or before (YYYY-MM-DD).
- `limit` (integer): Max rows (default 25, cap 100).
- `naics` (string): NAICS code prefix(es), CSV. Example: '5415' covers all IT services.
- `new_since` (string): Sub-recipient's FIRST observed subaward on or after this date (YYYY-MM-DD).
- `new_subawardee` (string): true → only first-time subawardees: subs with NO prime award on record.
- `offset` (integer): Rows to skip, for paging.
- `prime_award_id` (string): Prime award PIID / FAIN, exact match — all subs under one prime.
- `prime_recipient` (string): Prime recipient name fragment. Example: 'lockheed'.
- `prime_uei` (string): Prime recipient Unique Entity Identifier (SAM UEI), exact match.
- `psc` (string): Product/Service Code(s), CSV — exact.
- `sort` (string): field:direction — amount | action_date. Default 'action_date:desc'.
- `state` (string): Place-of-performance state code(s), CSV.
- `sub_recipient` (string): Sub-recipient (subcontractor/subgrantee) name fragment.
- `sub_uei` (string): Sub-recipient Unique Entity Identifier (SAM UEI), exact match.
- `type` (string): contract | grant (sub-contract vs sub-grant). CSV for both.

### `trials_trials_search` (~386 tokens)

Which companies are running clinical trials, and at what stage? Search ClinicalTrials.gov interventional industry-sponsored studies by lead sponsor, phase, status, condition, intervention type, a US state with a site, first-posted date (since), start date, or completion date. Returns the sponsor (entity-resolved where possible), phase, status, conditions, enrollment, and dates — a developer target list for CRO, trial-software, and commercial-ops sellers. ClinicalTrials.gov public records. [price: $0.05/row]

Input parameters:

- `completed_after` (string): Study completion on/after this date (YYYY-MM-DD).
- `condition` (string): Indication / condition fragment.
- `entity_id` (string): Resolved sponsor employer entity UUID (pivots to business360).
- `intervention_type` (string): Intervention type, CSV (DRUG | DEVICE | BIOLOGICAL | ...).
- `limit` (integer): Max rows (default 25, cap 100).
- `offset` (integer): Rows to skip.
- `phase` (string): Trial phase(s), CSV (PHASE1 | PHASE2 | PHASE3 | PHASE4 | EARLY_PHASE1).
- `since` (string): First posted to the registry on/after this date (YYYY-MM-DD).
- `sort` (string): first_posted_date | start_date | completion_date | last_update_date :asc|:desc. Default first_posted_date:desc.
- `sponsor` (string): Lead sponsor company — suffix/punctuation-insensitive.
- `started_after` (string): Study start on/after this date (YYYY-MM-DD).
- `state` (string): Has a trial site in these US state code(s), CSV.
- `status` (string): Overall status, CSV (RECRUITING | COMPLETED | TERMINATED | ...).

### `trials_sites_search` (~273 tokens)

Where are clinical trials being run, and by whom? Search ClinicalTrials.gov trial SITES by facility name, city, US state, lead sponsor company, trial phase, trial status, or trial start date (since). Returns each facility with its trial's sponsor, phase, status, and start date — a site / facility target list for trial-site services, patient-recruitment, and lab vendors. ClinicalTrials.gov public records. [price: $0.05/row]

Input parameters:

- `city` (string): Site city fragment.
- `facility` (string): Facility / site name fragment.
- `limit` (integer): Max rows (default 25, cap 100).
- `offset` (integer): Rows to skip.
- `phase` (string): Trial phase(s), CSV (PHASE1 | PHASE2 | PHASE3 | PHASE4).
- `since` (string): Trial started on/after this date (YYYY-MM-DD).
- `sort` (string): start_date :asc|:desc. Default start_date:desc.
- `sponsor` (string): Lead sponsor company — suffix/punctuation-insensitive.
- `state` (string): Site US state code(s), CSV.
- `status` (string): Trial overall status, CSV (RECRUITING | COMPLETED | ...).

### `fsis_establishments_search` (~364 tokens)

Which meat, poultry, or egg plants operate in a market? Search the USDA FSIS Meat, Poultry & Egg Product Inspection Directory (every federally-inspected establishment) by company, state, city, activity (slaughter/processing), species, size, or inspection-grant date (since, for new plants). Returns the establishment (entity-resolved where possible), address, activities, and size — a food-safety seller's entire target universe, not just OSHA-hit plants. FSIS public-domain records. [price: $0.05/row]

Input parameters:

- `activity` (string): Activity fragment (Slaughter, Processing, Egg Product, ...).
- `city` (string): City fragment.
- `company` (string): Establishment/company name — suffix/punctuation-insensitive.
- `entity_id` (string): Resolved employer entity UUID (pivots to business360).
- `establishment_number` (string): FSIS establishment number(s), CSV (M=meat, P=poultry, G=egg, I=import).
- `limit` (integer): Max rows (default 25, cap 100).
- `offset` (integer): Rows to skip.
- `processing` (string): true = processing establishments only.
- `since` (string): Inspection grant date on/after this date (YYYY-MM-DD) — surfaces newly-granted plants.
- `size` (string): Very Small | Small | Large (CSV).
- `slaughter` (string): true = slaughter establishments only.
- `sort` (string): grant_date | establishment_name :asc|:desc. Default grant_date:desc.
- `species` (string): meat | poultry | egg (CSV).
- `state` (string): State code(s), CSV.

### `fsis_recalls_search` (~322 tokens)

Which food companies have an FSIS meat/poultry/egg recall? Search USDA FSIS recall and public-health-alert records by recalling firm, classification (Class I/II/III or Public Health Alert), reason (allergens, contamination, misbranding), state, active status, outbreak link, or recall date (since). Returns the recalling firm (entity-resolved where possible), classification, reason, product, and dates — a remediation, sanitation, and displacement trigger for food-safety sellers. [price: $0.05/row]

Input parameters:

- `active` (string): true = currently-active recall notices only.
- `classification` (string): Class I | Class II | Class III | Public Health Alert (CSV).
- `company` (string): Recalling firm — suffix/punctuation-insensitive.
- `entity_id` (string): Resolved employer entity UUID.
- `limit` (integer): Max rows (default 25, cap 100).
- `offset` (integer): Rows to skip.
- `outbreak` (string): true = recalls related to an outbreak.
- `reason` (string): Reason fragment (Unreported Allergens, Product Contamination, Misbranding, ...).
- `recall_type` (string): Active Recall | Closed Recall | Public Health Alert (CSV).
- `since` (string): Recall on/after this date (YYYY-MM-DD).
- `sort` (string): recall_date | last_modified_date :asc|:desc. Default recall_date:desc.
- `state` (string): Distribution state code(s), CSV.

### `uspto_patents_search` (~435 tokens)

Which companies just got a US patent, and in what technology? Search recent granted patents (PatentsView, CC BY 4.0) by assignee organization (the company), inventor, CPC subclass/section, patent type, or assignee state/country and grant-date window. new_assignee=true isolates patents from an org with its first-ever grant (emerging/stealth signal); flipped_inventor=true finds a patent whose assignee gained an inventor from a prior different assignee (talent move). Public grants, observational. [price: $0.05/row]

Input parameters:

- `assignee` (string): Assignee organization (the company) name fragment.
- `assignee_entity` (string): Resolved entities.employers entity_id (UUID), exact match.
- `country` (string): Assignee country code(s), CSV (ISO).
- `cpc_section` (string): CPC section letter(s), CSV (A-H, Y).
- `cpc_subclass` (string): CPC subclass/class/section prefix(es), CSV. G06N=AI/ML, H04L=networking.
- `flipped_inventor` (string): true → only patents whose assignee gained an inventor from a prior different assignee.
- `granted_after` (string): Grant date on or after (YYYY-MM-DD).
- `granted_before` (string): Grant date on or before (YYYY-MM-DD).
- `inventor` (string): Inventor name fragment (last name or full name).
- `limit` (integer): Max rows (default 25, cap 100).
- `new_assignee` (string): true → only patents whose assignee org is new/emerging (first-ever grant recent).
- `offset` (integer): Rows to skip, for paging.
- `patent_type` (string): utility | design | plant | reissue, CSV.
- `sort` (string): field:direction — patent_date | num_claims. Default 'patent_date:desc'.
- `state` (string): Assignee state code(s), CSV (US assignees).

### `uspto_assignees_search` (~393 tokens)

Which companies are new to patenting, and in what technology? Search patent assignee ORGANIZATIONS (entity-resolved), each with first/last grant date, patent count, distinct CPC subclasses, and inventor count. new_assignee=true isolates orgs with their first-ever US patent grant (emerging/stealth signal for VC deal sourcing and corp-dev); narrow by CPC subclass, state/country, first-grant window, or patent_count_min. PatentsView grants (CC BY 4.0), observational. [price: $0.05/row]

Input parameters:

- `assignee` (string): Assignee organization (company) name fragment.
- `assignee_entity` (string): Resolved entities.employers entity_id (UUID), exact match.
- `country` (string): Assignee country code(s), CSV (ISO).
- `cpc_subclass` (string): CPC subclass/class/section the org patents in, CSV. G06N=AI/ML.
- `first_grant_after` (string): Org's first observed grant on or after (YYYY-MM-DD).
- `first_grant_before` (string): Org's first observed grant on or before (YYYY-MM-DD).
- `limit` (integer): Max rows (default 25, cap 100).
- `new_assignee` (string): true → only new/emerging assignees (first-ever patent grant recent).
- `offset` (integer): Rows to skip, for paging.
- `patent_count_max` (integer): Maximum granted-patent count held by the org.
- `patent_count_min` (integer): Minimum granted-patent count held by the org.
- `sort` (string): field:direction — patent_count | first_grant_date | last_grant_date. Default 'last_grant_date:desc'.
- `state` (string): Assignee state code(s), CSV.

### `msha_mines_search` (~443 tokens)

Which mines and quarries operate in a market, and who runs them? Search the MSHA Mines data set — every coal and metal/non-metal mine under MSHA jurisdiction (NAICS-21 operators that are OSHA-exempt and invisible elsewhere) — by operator, mine name, state, county, commodity, canvass, coal_metal, mine type, status, active-only, or status date (since). Returns the mine, its operator (entity-resolved) and controller, commodity, employees, and status. MSHA public-domain records. [price: $0.05/row]

Input parameters:

- `active` (string): true = currently-active mines only.
- `canvass` (string): Primary canvass (Coal | M/NM | Metal | Nonmetal | Sand & gravel | Stone), CSV.
- `coal_metal` (string): C = coal, M = metal/non-metal.
- `commodity` (string): Commodity/SIC or canvass fragment.
- `controller` (string): Controlling entity (parent) name fragment.
- `county` (string): County name fragment.
- `entity_id` (string): Resolved operator entity UUID (pivots to business360).
- `limit` (integer): Max rows (default 25, cap 100).
- `mine_id` (string): MSHA mine id(s), CSV.
- `mine_name` (string): Mine name fragment.
- `mine_type` (string): Surface | Underground | Facility (CSV).
- `offset` (integer): Rows to skip.
- `operator` (string): Mine operator/company name — suffix/punctuation-insensitive.
- `operator_id` (string): MSHA operator id(s), CSV.
- `since` (string): Current-status date on/after this date (YYYY-MM-DD) — surfaces newly-statused mines.
- `sort` (string): status_dt | mine_name | employees :asc|:desc. Default status_dt:desc.
- `state` (string): State code(s), CSV.
- `status` (string): Mine status (Active | Abandoned | Intermittent | Temporarily Idled | NonProducing), CSV.

### `msha_violations_search` (~473 tokens)

Which mine operators have MSHA citations or orders? Search the MSHA Violations data set (citations & orders at coal & M/NM mines) by cited company, mine id, state, cited 30 CFR standard (part_section), Mine Act section, citation/order type, health/safety area, negligence, likelihood, S&S (sig_sub), contested, year, or issue date (since). Returns the cited violator (entity-resolved), mine, standard cited, penalty, and dates. Reported exactly as cited, not a safety judgment. [price: $0.05/row]

Input parameters:

- `cit_ord` (string): Citation | Order | Safeguard (CSV).
- `coal_metal` (string): C = coal, M = metal/non-metal.
- `company` (string): Cited violator (operator/contractor) name — suffix/punctuation-insensitive.
- `contested` (string): true = contested citations only.
- `enforcement_area` (string): Health/safety hazard designation fragment.
- `entity_id` (string): Resolved violator entity UUID.
- `likelihood` (string): Accident likelihood (Highly | Reasonably | Unlikely | Occurred | NoLikelihood), CSV.
- `limit` (integer): Max rows (default 25, cap 100).
- `mine_id` (string): MSHA mine id(s), CSV.
- `negligence` (string): Negligence degree (CSV).
- `offset` (integer): Rows to skip.
- `part_section` (string): Cited 30 CFR part/section fragment (the standard cited).
- `section_of_act` (string): Mine Act section fragment.
- `sig_sub` (string): true = Significant & Substantial (S&S) citations only.
- `since` (string): Issued on/after this date (YYYY-MM-DD).
- `sort` (string): issue_date | penalty | affected :asc|:desc. Default issue_date:desc.
- `state` (string): State code(s), CSV (from the mine).
- `violator_id` (string): MSHA violator/operator id(s), CSV.
- `violator_type` (string): Operator | Contractor | Agent | Miner (CSV).
- `year` (string): Calendar year of issuance.

### `fda_warningletters_search` (~358 tokens)

Which FDA-regulated firms have an open FDA warning letter? Search the FDA Warning Letters index by company, product area (drugs, devices, biologics, food, dietary-supplement, cosmetics, tobacco, veterinary), issuing office, subject (CGMP, adulterated, misbranded), open/closed status, or issue/posted date. Returns the recipient firm (entity-resolved where possible), issuing office, subject, and response/close-out dates — a documented compliance gap and remediation trigger. [price: $0.05/row]

Input parameters:

- `closed` (string): false = open letters (no close-out) only; true = closed-out only.
- `company` (string): Recipient company name — suffix/punctuation-insensitive.
- `entity_id` (string): Resolved employer entity UUID (pivots to business360).
- `issuing_office` (string): FDA center / district office fragment.
- `limit` (integer): Max rows (default 25, cap 100).
- `offset` (integer): Rows to skip.
- `posted_since` (string): Posted date on/after this date (YYYY-MM-DD) — newly-published letters.
- `product_area` (string): drugs | devices | biologics | food | dietary-supplement | cosmetics | tobacco | veterinary | other (CSV).
- `responded` (string): true = a firm response letter is on record; false = none.
- `since` (string): Letter issue date on/after this date (YYYY-MM-DD).
- `sort` (string): letter_issue_date | posted_date :asc|:desc. Default letter_issue_date:desc.
- `subject` (string): Subject / alleged-violation fragment.

### `epa_rcra_handlers_search` (~368 tokens)

Which hazardous-waste handlers operate in a market? Search EPA RCRAInfo (every registered hazardous-waste generator, transporter, and treatment/storage/disposal facility) by company, state, city, generator status (LQG/SQG/VSQG), NAICS industry, or handler role (transporter, TSDF, active). Returns the handler (entity-resolved where possible), address, generator status, and role flags — an EHS vendor's entire regulated universe, not just ECHO hits. EPA public-domain records. [price: $0.05/row]

Input parameters:

- `active` (string): true = currently-active handler sites only.
- `city` (string): City fragment.
- `company` (string): Handler/company name — suffix/punctuation-insensitive.
- `entity_id` (string): Resolved employer entity UUID (pivots to business360).
- `generator_status` (string): LQG | SQG | VSQG | Non-Generator (CSV).
- `handler_id` (string): EPA RCRA handler ID(s), CSV.
- `limit` (integer): Max rows (default 25, cap 100).
- `lqg` (string): true = large-quantity generators only.
- `naics` (string): NAICS prefix(es), CSV (e.g. 32 = manufacturing, 3251 = basic chemicals).
- `offset` (integer): Rows to skip.
- `sort` (string): facility_name | generator_status :asc|:desc. Default facility_name:asc.
- `state` (string): State code(s), CSV.
- `transporter` (string): true = hazardous-waste transporters only.
- `tsdf` (string): true = operating treatment/storage/disposal facilities only.

### `epa_tri_search` (~355 tokens)

Which facilities report toxic chemical releases? Search the EPA Toxics Release Inventory (TRI Basic Data) by facility or parent company, state, chemical, CAS number, NAICS, reporting year, chemical class (carcinogen, PBT, PFAS), or minimum release amount. Returns the facility (entity-resolved where possible), parent company, chemical, and on/off-site release totals — the emissions universe for environmental, remediation, and EHS sellers. EPA public-domain; amounts self-reported. [price: $0.05/row]

Input parameters:

- `carcinogen` (string): true = OSHA-carcinogen chemicals only.
- `cas` (string): CAS number(s), CSV.
- `chemical` (string): Chemical name fragment.
- `company` (string): Facility or parent-company name — suffix/punctuation-insensitive.
- `entity_id` (string): Resolved employer entity UUID.
- `facility` (string): Facility name fragment.
- `limit` (integer): Max rows (default 25, cap 100).
- `min_total_releases` (string): Minimum total on+off-site release amount (reported unit, usually pounds).
- `naics` (string): NAICS prefix(es), CSV.
- `offset` (integer): Rows to skip.
- `pbt` (string): true = persistent bioaccumulative toxic (PBT) chemicals only.
- `pfas` (string): true = PFAS chemicals only.
- `sort` (string): total_releases | tri_year | facility_name :asc|:desc. Default total_releases:desc.
- `state` (string): State code(s), CSV.
- `year` (string): TRI reporting year(s), CSV.

### `nlrb_cases_search` (~392 tokens)

Which employers have NLRB labor activity — a union petition or an unfair-labor-practice charge? Search NLRB cases by employer, case type (CA/CB/RC/RM/...), category (ulp vs representation), status, region, state, union, election result, or filing date (since). Returns the employer (entity-resolved where possible), region, union, and — for elections — the tally. Cases as filed (a charge is an allegation, not a judgment); a labor-relations trigger for PEO/HR, benefits, staffing, and outplacement. [price: $0.05/row]

Input parameters:

- `case_category` (string): ulp | representation | unit (CSV).
- `case_type` (string): NLRB case type code(s), CSV (CA, CB, RC, RM, RD, UC, ...).
- `company` (string): Employer name — suffix/punctuation-insensitive.
- `entity_id` (string): Resolved employer entity UUID (pivots to business360).
- `has_election` (string): true = cases with an election tally on record.
- `limit` (integer): Max rows (default 25, cap 100).
- `offset` (integer): Rows to skip.
- `region` (string): NLRB region code(s), CSV (two-digit, e.g. 10).
- `result` (string): union | no union (election outcome, CSV).
- `since` (string): Filed on/after this date (YYYY-MM-DD).
- `sort` (string): date_filed | tally_issued_date | num_employees :asc|:desc. Default date_filed:desc.
- `state` (string): Employer state code(s), CSV.
- `status` (string): Open | Closed (CSV).
- `union` (string): Union / petitioner name fragment.

### `ttb_permittees_search` (~390 tokens)

Which distilleries, wineries, alcohol importers, and wholesalers hold a federal TTB basic permit in a market? Search the TTB List of Permittees by owner, operating/DBA name, permit number, permit_type (S/W/I/P), industry, state, city, county, zip, entity_id, or new-permit-only (new=true, a new-business trigger). Returns the permittee, operating name, address, industry type, and entity-resolved owner. Excludes brewers/tobacco (IRC 6103). TTB public-domain records. [price: $0.05/row]

Input parameters:

- `city` (string): Premise city fragment.
- `county` (string): Premise county fragment.
- `entity_id` (string): Resolved owner entity UUID (pivots to business360).
- `industry` (string): Industry type (Distilled Spirits Plant | Wine Producer | Importer (Alcohol) | Wholesaler (Alcohol)), CSV.
- `limit` (integer): Max rows (default 25, cap 100).
- `new` (string): true = newly-issued permits only (< 7 days) — a new-business trigger.
- `offset` (integer): Rows to skip.
- `operating_name` (string): Operating / DBA name fragment.
- `owner` (string): Permit-holder (owner) name — suffix/punctuation-insensitive.
- `permit_number` (string): TTB permit number(s), CSV (STATE-TYPE-SEQ).
- `permit_type` (string): S = distilled spirits, W = wine, I = importer, P = wholesaler.
- `sort` (string): permit_number | owner | city :asc|:desc. Default permit_number:asc.
- `state` (string): Premise state code(s), CSV.
- `zip` (string): Premise ZIP code.

### `hrsa_hpsa_search` (~342 tokens)

Where has the federal government designated a health-professional shortage? Search HRSA HPSAs by state, county, discipline (Primary Care | Dental Health | Mental Health), designation type, status, minimum HPSA score (0-26, higher = worse), population type, rural-only, or designated-since date. Returns the designation, score, geography, provider-to-population ratio, and shortage FTE — the buying moment for clinician-staffing, telehealth, and workforce vendors. [price: $0.05/row]

Input parameters:

- `county` (string): County name fragment.
- `designation_type` (string): Designation type fragment.
- `discipline` (string): Primary Care | Dental Health | Mental Health (CSV).
- `hpsa_id` (string): Exact HPSA ID.
- `limit` (integer): Max rows (default 25, cap 100).
- `min_score` (string): Minimum HPSA score (0-26, higher = worse).
- `offset` (integer): Rows to skip.
- `population_type` (string): Population-type fragment (Low Income, Migrant Farmworker...).
- `rural` (string): true = rural / partially-rural only; false = non-rural.
- `since` (string): Designated on or after this date (YYYY-MM-DD).
- `sort` (string): score | designation_date | shortage_fte :asc|:desc. Default score:desc.
- `state` (string): Primary state/territory code(s), CSV.
- `status` (string): Designated | Proposed for Withdrawal | Withdrawn (CSV).

### `hrsa_mua_search` (~267 tokens)

Which areas and populations has HRSA designated Medically Underserved (MUA/P)? Search by state, county, status, designation type, MUA/P id, maximum IMU score (0-100, LOWER = more underserved), or designated-since date. Returns the service area, IMU score, geography, and dates — the safety-net / FQHC territory for staffing, telehealth, and access programs. HRSA public-domain records. [price: $0.05/row]

Input parameters:

- `county` (string): County name fragment.
- `designation_type` (string): Designation type fragment.
- `limit` (integer): Max rows (default 25, cap 100).
- `max_imu` (string): Maximum IMU score (0-100, LOWER = more underserved).
- `mua_id` (string): Exact MUA/P ID.
- `offset` (integer): Rows to skip.
- `since` (string): Designated on or after this date (YYYY-MM-DD).
- `sort` (string): imu_score | designation_date :asc|:desc. Default imu_score:asc.
- `state` (string): State/territory code(s), CSV.
- `status` (string): Designated | Withdrawn (CSV).

### `cms_timely_care_search` (~291 tokens)

How is a hospital's emergency department performing on CMS timely-care measures? Search CMS 'Timely and Effective Care - Hospital' by state, facility_id (CCN), measure (OP_18b median ED time, OP_22 left-without-being-seen, SEP_1...), condition, city, min/max score, reporting-period end, or latest_only. Returns per-hospital score, sample, measure period, and snapshot. CMS Care Compare public-domain records. [price: $0.05/row]

Input parameters:

- `city` (string): City fragment.
- `condition` (string): Condition fragment.
- `facility_id` (string): CMS CCN(s), CSV.
- `latest_only` (string): true = only each facility×measure's most recent period.
- `limit` (integer): Max rows (default 25, cap 100).
- `max_score` (string): Maximum numeric score.
- `measure` (string): Measure id(s), CSV (OP_18b, OP_22, SEP_1...).
- `min_score` (string): Minimum numeric score.
- `offset` (integer): Rows to skip.
- `period_end` (string): Exact reporting-period end date (YYYY-MM-DD).
- `sort` (string): score | sample | period_end :asc|:desc.
- `state` (string): Hospital state code(s), CSV.

### `cms_timely_care_delta` (~266 tokens)

Which hospitals' emergency-department throughput is degrading period-over-period? Windows the CMS Timely & Effective Care snapshots to compute each hospital's latest-vs-prior change on a measure (default OP_18b median ED time; higher = worse). Filter by measure, state, facility_id, direction (worse | better | any), min_pct_change, min_delta, and min_sample. Returns latest_score, prior_score, delta, and pct_change — the founder's ED-staffing-pressure tell. CMS Care Compare public-domain records. [price: $0.05/row]

Input parameters:

- `direction` (string): worse (default) | better | any.
- `facility_id` (string): CMS CCN(s), CSV.
- `limit` (integer): Max rows (default 25, cap 100).
- `measure` (string): Measure id to compare across periods (default OP_18b).
- `min_delta` (string): Minimum absolute change in the chosen direction.
- `min_pct_change` (string): Minimum period-over-period % change in the chosen direction.
- `min_sample` (string): Minimum latest-period sample size.
- `offset` (integer): Rows to skip.
- `state` (string): Hospital state code(s), CSV.

### `hcris_cost_reports_search` (~238 tokens)

What does a hospital's Medicare cost report show for staffing? Search the flattened HCRIS S-3 (CMS 'Hospital Provider Cost Report') by state, CCN, name, fiscal year, control_type, or min_beds. Returns FTE employees, beds, discharges, salaries, contract labor, plus computed FTE-per-bed and FTE-per-1k-discharges staffing ratios. CMS public-domain records. [price: $0.05/row]

Input parameters:

- `ccn` (string): Provider CCN(s), CSV.
- `control_type` (string): Type of control fragment.
- `limit` (integer): Max rows (default 25, cap 100).
- `min_beds` (string): Minimum number of beds.
- `name` (string): Hospital name fragment.
- `offset` (integer): Rows to skip.
- `sort` (string): fte_employees | number_of_beds | total_discharges | fiscal_year :asc|:desc.
- `state` (string): Hospital state code(s), CSV.
- `year` (string): Fiscal year(s), CSV.

### `hcris_staffing_decline_search` (~274 tokens)

Which hospitals cut staffing year-over-year? Compares each CCN's latest HCRIS fiscal year to the prior on a staffing ratio (metric=fte_per_bed default, or fte_per_discharge) and returns those whose ratio DROPPED. Filter by state, min_pct_drop, year, ccn, name, control_type, min_beds. Returns latest/prior ratio, delta and pct_change — an understaffing tell. CCN joins CMS timely-care. CMS public-domain records. [price: $0.05/row]

Input parameters:

- `ccn` (string): Provider CCN(s), CSV.
- `control_type` (string): Type of control fragment.
- `limit` (integer): Max rows (default 25, cap 100).
- `metric` (string): fte_per_bed (default) | fte_per_discharge.
- `min_beds` (string): Minimum beds in the latest year.
- `min_pct_drop` (string): Minimum YoY % drop in the ratio.
- `name` (string): Hospital name fragment.
- `offset` (integer): Rows to skip.
- `state` (string): Hospital state code(s), CSV.
- `year` (string): Restrict the LATEST fiscal year(s) compared, CSV.

### `fcc_licenses_search` (~427 tokens)

Who just got a new or modified FCC wireless license (a buildout tell)? Search the FCC ULS license transactions (microwave, land-mobile private/commercial, paging, broadcast-aux) by licensee company, call sign, radio_service, service_group, state, city, status, applicant_type, FRN, grant date (granted_since), or last-action date (since). Returns the license, its entity-resolved licensee, radio service, status, and dates. FCC public-domain records. [price: $0.05/row]

Input parameters:

- `applicant_type` (string): Applicant type code(s), CSV (B business, C corp, G govt, I individual, L limited-liability, P partnership).
- `call_sign` (string): Call sign(s), CSV.
- `city` (string): Premise city fragment.
- `company` (string): Licensee company/agency name — suffix/punctuation-insensitive.
- `entity_id` (string): Resolved licensee entity UUID (pivots to business360).
- `frn` (string): FCC Registration Number(s), CSV.
- `granted_since` (string): Grant date on/after this date (YYYY-MM-DD) — surfaces new license grants.
- `limit` (integer): Max rows (default 25, cap 100).
- `offset` (integer): Rows to skip.
- `radio_service` (string): FCC radio service code(s), CSV (e.g. MW microwave, IG industrial/business, PW public safety).
- `service_group` (string): microwave | land-mobile-private | land-mobile-commercial | paging | land-mobile-broadcast-aux (CSV).
- `since` (string): Last-action date on/after this date (YYYY-MM-DD) — recent transactions.
- `sort` (string): grant_date | last_action_date | call_sign :asc|:desc. Default last_action_date:desc.
- `state` (string): Premise state code(s), CSV.
- `status` (string): License status code(s), CSV (A = active).

### `faa_aircraft_search` (~412 tokens)

Who owns which aircraft, and what did they just register? Search the FAA aircraft registry by registrant/owner, N-number, manufacturer, model, type_aircraft, type_registrant (1 individual/3 corp/7 LLC), state, city, status, fractional-only, or certificate-issue date (registered_since — new-ownership tell). Returns the aircraft (make/model/type/year/seats) and its entity-resolved registrant. FAA public-domain records. [price: $0.05/row]

Input parameters:

- `city` (string): Registrant city fragment.
- `entity_id` (string): Resolved registrant entity UUID (pivots to business360).
- `fractional` (string): true = fractional-ownership aircraft only.
- `limit` (integer): Max rows (default 25, cap 100).
- `mfr` (string): Manufacturer name fragment.
- `model` (string): Model name fragment.
- `n_number` (string): N-number(s), CSV (leading N optional).
- `offset` (integer): Rows to skip.
- `registered_since` (string): Certificate-issue date on/after this date (YYYY-MM-DD) — new registrations/ownership.
- `registrant` (string): Registrant/owner name — suffix/punctuation-insensitive.
- `since` (string): Last-action date on/after this date (YYYY-MM-DD).
- `sort` (string): cert_issue_date | last_action_date | year_mfr :asc|:desc. Default cert_issue_date:desc.
- `state` (string): Registrant state code(s), CSV.
- `status` (string): Registration status code(s), CSV (V = valid).
- `type_aircraft` (string): Type-aircraft code(s), CSV (4 fixed-wing single, 5 fixed-wing multi, 6 rotorcraft).
- `type_registrant` (string): Registrant-type code(s), CSV (1 individual, 3 corporation, 7 LLC).

### `faa_deregistrations_search` (~286 tokens)

Which aircraft were just deregistered (sold, exported, scrapped, or repossessed)? Search FAA aircraft deregistrations (cancel date >= 2022) by last registrant/owner, N-number, mailing state, status, or cancel date (since/before) — the asset-disposal / ownership-exit event for auctioneers, dealers, and asset-based lenders. Returns the aircraft, its last entity-resolved registrant, and the cancel date. FAA public-domain records. [price: $0.05/row]

Input parameters:

- `before` (string): Cancel date on/before this date (YYYY-MM-DD).
- `entity_id` (string): Resolved registrant entity UUID.
- `limit` (integer): Max rows (default 25, cap 100).
- `n_number` (string): N-number(s), CSV (leading N optional).
- `offset` (integer): Rows to skip.
- `registrant` (string): Last registrant/owner name — suffix/punctuation-insensitive.
- `since` (string): Cancel date on/after this date (YYYY-MM-DD).
- `sort` (string): cancel_date | last_action_date :asc|:desc. Default cancel_date:desc.
- `state` (string): Mailing state code(s), CSV.
- `status` (string): Status code(s), CSV.

### `phmsa_operators_search` (~251 tokens)

Which pipeline operators are in a market, and who are they? Search the PHMSA pipeline operator roster by operator name, operator_id (OPID), HQ state, operator_type, or commodity. Returns the operator, HQ address, commodity, type, and entity-resolved company — the target universe for oil-and-gas services, pipeline-integrity & environmental consultants, energy insurers, and industrial real estate. PHMSA public-domain records. [price: $0.05/row]

Input parameters:

- `commodity` (string): Commodity fragment.
- `entity_id` (string): Resolved operator entity UUID (pivots to business360).
- `limit` (integer): Max rows (default 25, cap 100).
- `offset` (integer): Rows to skip.
- `operator` (string): Pipeline operator name — suffix/punctuation-insensitive.
- `operator_id` (string): PHMSA operator id (OPID)(s), CSV.
- `operator_type` (string): Operator/ownership type fragment.
- `sort` (string): operator_name | report_year :asc|:desc. Default operator_name:asc.
- `state` (string): HQ state code(s), CSV.

### `phmsa_incidents_search` (~308 tokens)

Which pipeline operators had a reported incident? Search PHMSA pipeline incidents (2010+, gas distribution / hazardous liquid / gas transmission) by operator, operator_id, mode, state, cause, commodity, minimum cost (min_cost), year, or incident date (since). Returns the incident, its entity-resolved operator, location, commodity released, casualties, cost, and cause — a remediation / integrity / claims trigger. Reported exactly as recorded, not a safety judgment. [price: $0.05/row]

Input parameters:

- `cause` (string): Incident cause fragment.
- `commodity` (string): Commodity released fragment.
- `entity_id` (string): Resolved operator entity UUID.
- `limit` (integer): Max rows (default 25, cap 100).
- `min_cost` (string): Minimum total cost (USD).
- `mode` (string): Pipeline mode (gas_distribution | hazardous_liquid | gas_transmission), CSV.
- `offset` (integer): Rows to skip.
- `operator` (string): Operator name — suffix/punctuation-insensitive.
- `operator_id` (string): PHMSA operator id(s), CSV.
- `since` (string): Incident date on/after this date (YYYY-MM-DD).
- `sort` (string): incident_date | total_cost | fatalities :asc|:desc. Default incident_date:desc.
- `state` (string): Incident state code(s), CSV.
- `year` (string): Incident year.

### `ofac_screen_sanctions` (~219 tokens)

Is this company on a U.S. sanctions or export-control list? One-call screen of the Consolidated Screening List — OFAC SDN, BIS Entity/Denied-Persons/Unverified/Military-End-User, State DTC/ISN, Treasury SSI/CMIC/PLC/NS-MBS — by normalized name and every published AKA (optional list= filter). Returns an attested records_found verdict, per-list hit counts, and the matching listings with program, type, and addresses. A name match is a screening hit to adjudicate, not a determination. [price: $0.1/call]

Input parameters:

- `list` (string): Optional source-list code to narrow the screen (SDN | EL | DPL | DTC | SSI | UVL | ISN | PLC | MEU | CMIC | NS-MBS | CAP).
- `name` (string, required): Company/party name to screen — legal suffixes and punctuation normalized; matched against the primary name and every alternate name/AKA.

### `sba_loans_search` (~424 tokens)

Which companies just got SBA growth capital — a 7(a) guaranty or 504/CDC debenture? Search the SBA 7(a) & 504 FOIA loan-approval roster by company, state, project_state, naics (prefix), program (7A/504), amount_min/amount_max, since (approved on/after a date — the funded-to-grow trigger), business_type, franchise, lender/CDC, entity_id. Returns the borrower, address, gross approval amount, approval date/FY, NAICS, funding lender, and entity-resolved borrower. SBA public-domain records. [price: $0.05/row]

Input parameters:

- `amount_max` (number): Maximum gross approval amount (USD).
- `amount_min` (number): Minimum gross approval amount (USD).
- `business_type` (string): Business type fragment (Corporation | Individual | Partnership).
- `company` (string): Borrower (company) name — suffix/punctuation-insensitive.
- `entity_id` (string): Resolved borrower entity UUID (pivots to business360).
- `franchise` (string): Franchise name fragment.
- `lender` (string): Funding lender fragment — 7(a) bank name / 504 CDC name.
- `limit` (integer): Max rows (default 25, cap 100).
- `naics` (string): NAICS code prefix(es), CSV (e.g. 23 = construction).
- `offset` (integer): Rows to skip.
- `program` (string): 7A = 7(a) guaranty, 504 = 504/CDC debenture.
- `project_state` (string): Project (place-of-use) state code(s), CSV.
- `since` (string): Approved on/after this date (YYYY-MM-DD) — the growth-capital recency window.
- `sort` (string): approval_date | amount | company | jobs :asc|:desc. Default approval_date:desc.
- `state` (string): Borrower state code(s), CSV.

### `list_industries` (~67 tokens)

Discover ParseData's industries and their curated, ready-made Parses (each with a plain-language ELI12 explanation and an id you can pass to build_parse/get_leads). FREE.

Input parameters:

- `industry` (string): An industry slug to return in full; omit for the whole list.

### `build_parse` (~350 tokens)

Compose a lead-generation Parse from what you sell and who you target. Returns the signals/legs it watches, a plain-English restatement, a per-Parse narrative, a fit score, the expected new-leads-per-period rate, and a small free sample. This is the FREE preview of get_leads.

Input parameters:

- `attributes` (array): Ideal-customer descriptors, e.g. ["small / emerging", "multi-location"].
- `buying_moments` (array): Structured pick: library template ids (buying-moment ids) from build_parse/list_industries output. Each maps to a real signal.
- `industry` (string): An industry slug from list_industries (e.g. "freight-factoring", "pharma-sales") — scopes/labels the pick. Not a selector on its own.
- `parse` (object): Alternatively, the `parse` object returned by build_parse — replays the exact composed intent so a preview and its paid pull match.
- `preset_id` (string): Alternatively, a curated-Parse or preset id (from list_industries) to compose a proven Parse.
- `query` (string): OPTIONAL value-prop LABEL only (what you sell) — it does NOT select the Parse and is never inferred. The Parse is composed ONLY from an explicit selection: a `preset_id`, or `target_entity` + `buying…
- `states` (array): Two-letter US state codes to scope to; empty = nationwide.
- `target_entity` (string): Structured pick: the entity you SELL TO. Taken verbatim (never guessed), so a pharma-brand seller can't be mis-targeted onto physicians.

### `get_leads` (~436 tokens)

Get the actual B2B leads for a Parse: the matching companies and the public-record events that flagged them (real rows, including any free native contacts on the source record). PAID, priced PER LEAD RETURNED — connect with your API key (metered key billed per lead, or a SaaS key covered by your plan's lead entitlement) or pay keyless with an x402 USDC payment on Base or Solana (prepaid for up to `limit` leads). Compose the Parse with build_parse first, then pass its `parse` here.

Input parameters:

- `attributes` (array): Ideal-customer descriptors, e.g. ["small / emerging", "multi-location"].
- `buying_moments` (array): Structured pick: library template ids (buying-moment ids) from build_parse/list_industries output. Each maps to a real signal.
- `industry` (string): An industry slug from list_industries (e.g. "freight-factoring", "pharma-sales") — scopes/labels the pick. Not a selector on its own.
- `limit` (integer): Rows per page (default 25, max 100).
- `offset` (integer): Row offset for paging (default 0).
- `parse` (object): Alternatively, the `parse` object returned by build_parse — replays the exact composed intent so a preview and its paid pull match.
- `preset_id` (string): Alternatively, a curated-Parse or preset id (from list_industries) to compose a proven Parse.
- `query` (string): OPTIONAL value-prop LABEL only (what you sell) — it does NOT select the Parse and is never inferred. The Parse is composed ONLY from an explicit selection: a `preset_id`, or `target_entity` + `buying…
- `states` (array): Two-letter US state codes to scope to; empty = nationwide.
- `target_entity` (string): Structured pick: the entity you SELL TO. Taken verbatim (never guessed), so a pharma-brand seller can't be mis-targeted onto physicians.

### `enrich_contact` (~171 tokens)

Reveal the decision-maker at a target company (named person, title, work email, LinkedIn), third-party enriched. PAID: connect with your API key (a SaaS key spends 1 enrichment credit; a metered key is billed per reveal) or pay keyless via an x402 USDC payment. Never returns a government record; always source:enriched.

Input parameters:

- `company` (string, required): Target company / employer / carrier legal name. Required.
- `domain` (string): Company web domain — the single strongest match hint.
- `name` (string): A specific person name, if already known.
- `state` (string): State/region hint to disambiguate common names.
- `title` (string): Desired role/seniority, e.g. "VP Operations", "owner".

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/io-govparse-gateway/api#diagnostics

## Score history

- 2026-08-03: 67
- 2026-08-02: 66
- 2026-08-01: 66
- 2026-07-31: 66
- 2026-07-30: 64
- 2026-07-29: 63
- 2026-07-28: 63
- 2026-07-27: 62
- 2026-07-26: 60

## Links

- Remote endpoint: https://api.govparse.io/mcp
- Website: https://govparse.io/
- Changelog RSS feed: https://verifymcp.io/servers/io-govparse-gateway/api/changelog.xml
- Changelog JSON feed: https://verifymcp.io/servers/io-govparse-gateway/api/changelog.json
- HTML version of this page: https://verifymcp.io/servers/io-govparse-gateway/api