Skip to main content
Glama
whetstone-mcp

whetstone-mcp

Official

whetstone-mcp

An MCP server for U.S. public-records data — so AI agents and MCP clients (Claude Desktop, Cursor, etc.) can look up business records, screen names against government watchlists, and pull federal awards directly.

Powered by the Whetstone actors on Apify. All data is official U.S. government public-record data.

Tools

Tool

What it does

business_search

Official Secretary of State business registration (KYB) across 25 states

new_business_filings

Newly registered businesses from 10 states, windowed by date

watchlist_screen

Screen a name against 12 federal watchlists (OFAC, BIS, State Dept)

federal_awards

A company's federal contracts, grants, and loans (USAspending.gov)

Related MCP server: PubRecords MCP

Setup

You need a free Apify API token (apify.com → Settings → API & Integrations). Runs bill to your Apify account under pay-per-result pricing (~$2 per 1,000 result rows; the free tier covers light use).

Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "whetstone": {
      "command": "npx",
      "args": ["-y", "whetstone-mcp"],
      "env": { "APIFY_TOKEN": "apify_api_your_token_here" }
    }
  }
}

Restart Claude Desktop. The four tools then appear and the agent can call them.

Any MCP client

Run the server over stdio:

APIFY_TOKEN=apify_api_... npx -y whetstone-mcp

Notes & limits

  • Heavy queries can be slow. Business search across all 25 states (or filings across all 10) runs synchronously and may approach the 300-second limit. Pass a states subset for faster, cheaper runs.

  • Name-based matching. Watchlist and federal-award results are matched by name and are not identity confirmation — verify a hit against the official source before acting. Nothing here is legal, compliance, or financial advice.

  • A free, interactive version of the combined lookup is at whetstonetools.com/company-check, and there's also an n8n community node.

License

MIT © Whetstone Tools · support@whetstonetools.com

Available Tools

4 tools
federal_awardsA

Look up a U.S. company's federal contracts, grants, and loans (USAspending.gov) with amounts, agencies, and dates. Useful for B2G sales intel and due diligence.

Inputs: recipientName (required), awardTypes, maxResults, sinceFiscalYear.

new_business_filingsA

Pull newly registered U.S. businesses from official state sources (10 states), windowed by date. Useful for sales-lead feeds and monitoring brand-new companies.

Inputs: states, daysBack, sinceDate, maxResultsPerState.

watchlist_screenA

Screen a person or business name against 12 U.S. government watchlists (OFAC SDN, BIS Entity/Denied/Unverified/MEU, State Dept Debarred/ISN, and more) via the Consolidated Screening List. Returns fuzzy-matched records. NOTE: name-based matching, NOT identity confirmation — verify any hit against the official source before acting.

Inputs: name (required), lists, minScore, maxResults, +1 more.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A4/5.0

    Scored across 4 tools

    Disambiguation5/5

    Each tool targets a distinct data source and workflow: registration lookup, new-filing lead generation, watchlist screening, and federal awards. business_search and new_business_filings both involve state registrations, but one is a name-based lookup and the other is a date-windowed list, so there is little risk of misselection.

    Naming Consistency3/5

    All names are lowercase snake_case and readable, but they do not follow a single verb_noun or object_verb pattern. business_search and watchlist_screen end with action-like words, while new_business_filings and federal_awards are pure noun phrases, creating a mixed but not chaotic convention.

    Tool Count5/5

    Four tools is a tight, well-scoped set for a business-intelligence server. Each tool covers a distinct area without redundancy, and none feel like filler or unnecessary splits.

    Completeness4/5

    The set covers the core U.S. business-intelligence workflows: entity lookup, new-company lead generation, sanctions/debarment screening, and federal award data. Minor additions like beneficial-ownership details or state-level contract data would improve coverage, but the current surface is not incomplete enough to cause agent failures.

    Maintenance

    ActivityInactive
    ResponsivenessOutdated data

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      A
      quality
      D
      maintenance
      An MCP server that gives AI agents access to US business entity data, enabling searches across 9 state registries, SEC EDGAR filings, federal contracts, and lobbying disclosures.
      6
      1
      MIT
    • A
      license
      Not graded
      quality
      D
      maintenance
      MCP server that gives AI agents structured access to Public Procurement Registry data, allowing natural language queries for planned procurements, open tenders, contract awards, buyers, and winning suppliers.
      MIT
    • A
      license
      A
      quality
      D
      maintenance
      MCP server for querying US federal contracts and public company financial data from USAspending and SEC EDGAR. Enables searching expiring contracts, screening companies by metrics, and retrieving financial histories.
      4
      MIT