gov-data-mcp

license-verifier

Verify US professional licenses across state boards: status, expiration and disciplinary actions, with NPI cross-checks and HHS-OIG exclusion screening. Bulk roster screening supported.

Instructions

License Verification API — Nurses, MDs & OIG Exclusions. Primary source verification for US professional licenses. Search 19 state boards by name or license number: status, expiration, disciplinary actions. Cross-checks the NPPES NPI registry and screens the HHS-OIG exclusion list. Bulk roster screening. Newly licensed clinicians feed (roster-delta). Reads live from the official government source. Call describe-gov-data-tool (free) for a verified example input. COST AND SIDE EFFECTS: read-only with respect to the government source — it never writes to any external system — but each call starts a metered run on YOUR Apify account, billed $0.01 per result ($10 per 1,000). Lower on paid Apify plans, down to $3.00 per 1,000. Nothing is charged when a run fails. Store page: https://apify.com/malonestar/license-verifier

Input Schema

TableJSON Schema

Name	Required	Description
`city`	No	City to filter licensees by. Not every board publishes a city column.
`mode`	No	Leave empty for the classic lookup/roster behaviour. Set "roster-delta" for the newly-credentialed feed: every credential ORIGINALLY ISSUED in the last sinceDays on the selected boards, one row each, with NPI cross-walk and OIG screen. The first run seeds a named baseline and emits the window as event_type "inventory"; later runs emit only event_type "newly_licensed" (a credential absent from the baseline AND issued on/after the previous run minus 7 days). A run against unchanged data emits 0 rows and bills nothing. Boards with an original-issue date: WA (DOH), TX-BON, TX-LVN, TX-APRN, IL, CO, CT, DE, OR (CCB), WA-CPA, WA-CONTRACTOR, NY-NOTARY, NY-COS, NY-APPRAISER. In this mode maxResults is the TOTAL row cap for the run (newest credentials first).
`name`	No	Full-name search that works across every board (handles combined name fields). Use this if lastName/firstName return nothing.
`roster`	No	Batch mode: verify a whole roster of professionals in one run. Each item: {firstName, lastName, state (optional — omit to search every board), profession (optional), middleName (optional, improves scoring)}. Emits one verdict row per entry with verdict, match_score, match_tier, NPI cross-walk, OIG exclusion screen and board-action status. Capped at 200 entries per run. Every verdict row is billable, including NOT_FOUND and INCONCLUSIVE_SOURCE_ERROR — a verified negative is the deliverable.
`states`	No	State codes or explicit board IDs. A bare state code searches EVERY board in that state - e.g. "TX" covers TDLR trades AND the Board of Nursing (RN, LVN, APRN). Use a hyphenated ID to target ONE board: IL-IDFPR, CT-DCP, CO-DORA, TX-TDLR, TX-BON (RN), TX-LVN, TX-APRN, OR-CCB, OR-BCD, NY-RACING (horse racing only), NY-RE, NY-COS, NY-NOTARY, NY-APPRAISER, WA-DOH (health professions), WA-CPA, WA-CONTRACTOR, DE-DPR, VT-DFS. States: CO, CT, DE, IL, NY, OR, TX, VT, WA. Example: ["WA"].
`lastName`	No	Licensee last name (partial match). Example: "Threlkeld". Applied by default if omitted: "".
`firstName`	No	Licensee first name (partial match). Supplying it raises match confidence sharply — first + last name exact is the threshold for a confident verdict. Example: "Judson". Applied by default if omitted: "".
`npiLookup`	No	For each roster entry, look the person up in the federal NPI registry and use their self-reported state license number to pin down the exact board record. This is what turns 125 same-name candidates into one verified match, and it returns NPI, taxonomy and practice address. Applied by default if omitted: true.
`sinceDays`	No	roster-delta only. How many days back the original-issue-date window reaches (1-400). The seeding run emits this whole window as inventory; later runs emit only credentials issued since the previous run (minus a 7-day publication slack). A non-integer or out-of-range value fails the run before any request is made. Example: 30.
`maxResults`	No	Maximum number of license records to return per board. Each returned row is billable, so start small. Example: 10. Applied by default if omitted: 200.
`statusOnly`	No	Return only license number, type, status, expiration, provenance and the OIG exclusion flags. Handy for recurring renewal monitoring. NOTE: this is the SAME price per row as a full record — it returns less data, not cheaper data. Applied by default if omitted: false.
`licenseType`	No	e.g. "Registered Nurse", "Real Estate", "Cosmetology", "Professional Engineer". Boards without a license-type column skip this filter and say so in the log and in unsupported_filters.
`professions`	No	roster-delta only. Restrict to these professions, matched case-insensitively on the normalised profession (e.g. "Registered Nurse") or as a prefix of the board's raw credential type ("Registered Nurse" reaches "Registered Nurse License" and "Registered Nurse Temporary Practice Permit" but NOT "Advanced Registered Nurse Practitioner"). Examples: "Registered Nurse", "Licensed Practical Nurse", "Physician And Surgeon", "Dentist", "Pharmacist", "Physical Therapist". Omit for every profession the board publishes. Changing this filter starts a NEW delta baseline (the baseline is scoped by boards + professions). Example: ["Registered Nurse"].
`businessName`	No	Business or DBA name to search (partial match).
`licenseNumber`	No	Exact license number to verify. The most precise search available — use it when you have it.
`checkDiscipline`	No	Join the best-matching licensee against secondary board-action datasets: Delaware DPR disciplinary actions and the NYS Office of Professional Medical Conduct. A failed lookup is reported as unknown, never as 'no action on file'. Applied by default if omitted: true.
`onlyDisciplined`	No	Return only licensees with a disciplinary history. Honoured by IL-IDFPR, CO-DORA, DE-DPR, WA-DOH, TX-BON and TX-LVN. Target those board IDs directly rather than a bare state code, or sibling boards that publish no disciplinary column will also return rows (they are reported in unsupported_filters). Applied by default if omitted: false.
`socrataAppToken`	No	Optional free Socrata app token to raise rate limits.
`screenExclusions`	No	Check every result against the federal HHS-OIG List of Excluded Individuals/Entities (83,000+ records, refreshed monthly). Matched on NPI first, then last+first+state. A surname-only hit is NEVER reported as an exclusion — it is flagged for review instead. Adds no per-row cost. Applied by default if omitted: true.
`rosterLimitPerBoard`	No	How many candidate records to pull per board for each roster entry before scoring. Higher values reduce the chance of missing the right person for a common surname; candidates_truncated tells you when the cap was hit. Applied by default if omitted: 100.

Schema Changelog

Changes observed during successful MCP inspections.

Changed3 schema fields changedv1.2.0Sep 30, 2026

addedInput schema / properties / mode

Added value: +{
+  "description": "Leave empty for the classic lookup/roster behaviour. Set \"roster-delta\" for the newly-credentialed feed: every credential ORIGINALLY ISSUED in the last sinceDays on the selected boards, one row each, with NPI cross-walk and OIG screen. The first run seeds a named baseline and emits the window as event_type \"inventory\"; later runs emit only event_type \"newly_licensed\" (a credential absent from the baseline AND issued on/after the previous run minus 7 days). A run against unchanged data emits 0 rows and bills nothing. Boards with an original-issue date: WA (DOH), TX-BON, TX-LVN, TX-APRN, IL, CO, CT, DE, OR (CCB), WA-CPA, WA-CONTRACTOR, NY-NOTARY, NY-COS, NY-APPRAISER. In this mode maxResults is the TOTAL row cap for the run (newest credentials first).",
+  "enum": [
+    "roster-delta"
+  ],
+  "type": "string"
+}

addedInput schema / properties / professions

Added value: +{
+  "description": "roster-delta only. Restrict to these professions, matched case-insensitively on the normalised profession (e.g. \"Registered Nurse\") or as a prefix of the board's raw credential type (\"Registered Nurse\" reaches \"Registered Nurse License\" and \"Registered Nurse Temporary Practice Permit\" but NOT \"Advanced Registered Nurse Practitioner\"). Examples: \"Registered Nurse\", \"Licensed Practical Nurse\", \"Physician And Surgeon\", \"Dentist\", \"Pharmacist\", \"Physical Therapist\". Omit for every profession the board publishes. Changing this filter starts a NEW delta baseline (the baseline is scoped by boards + professions). Example: [\"Registered Nurse\"].",
+  "items": {
+    "type": "string"
+  },
+  "type": "array"
+}

addedInput schema / properties / sinceDays

Added value: +{
+  "description": "roster-delta only. How many days back the original-issue-date window reaches (1-400). The seeding run emits this whole window as inventory; later runs emit only credentials issued since the previous run (minus a 7-day publication slack). A non-integer or out-of-range value fails the run before any request is made. Example: 30.",
+  "maximum": 400,
+  "minimum": 1,
+  "type": "integer"
+}

First observedv1.0.2Aug 20, 2026

TDQS

A4.3/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Adds substantial context beyond the annotations: exact pricing ($0.01/result, $10 per 1,000, down to $3.00 on paid plans), that failed runs bill nothing, that each call starts a metered run on the caller's own Apify account, and rate-limit relief via socrataAppToken. It also reconciles the readOnlyHint=false annotation by clarifying the read-only property applies to the government source, not to the caller's account — a genuinely useful disclosure.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loads purpose, then cross-checks, then a clearly labelled COST AND SIDE EFFECTS block, which is the right ordering for a metered tool. Length is justified by the 20-parameter surface, though the store-page URL and the repeated marketing phrasing ('Primary source verification') are mild filler.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a complex, 20-param, no-output-schema tool the description covers scope, cost, side effects, cross-checks and mode behaviour well, and points to a helper tool for example inputs. It does not enumerate the return shape for the classic lookup path (output schema is absent), leaving a small gap, but the roster verdict fields are described.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100% and the per-parameter descriptions are extremely detailed (mode semantics, board IDs, delta baselines, billing per row), so the schema carries the parameter burden. The top-level description adds only marginal parameter meaning ('start small' on maxResults, pointer to describe-gov-data-tool), so baseline 3 is appropriate.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Primary source verification for US professional licenses'), names the exact scope (19 state boards, name or license number, status/expiration/discipline) and the federal cross-checks (NPPES NPI, HHS-OIG). The sibling tools are all in unrelated domains (EPA, FAA, FEMA, wetlands, parcels), so no differentiation ambiguity exists.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives clear context for the main modes (classic lookup/roster vs roster-delta newly-licensed feed) and routes the agent to describe-gov-data-tool for a verified example input. It also advises starting small on maxResults because rows are billable. It does not, however, state when to prefer this tool over an alternative approach or any explicit exclusions.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server

license-verifier

Instructions

Input Schema

Schema Changelog

TDQS

Other Tools