Kempt Horizon Data Toolkit

Web pages to clean Markdown, email/phone/CSV cleaning, and live SEC EDGAR, Federal Register, USAspending and Grants.gov data. 17 read-only tools, hosted, OAuth.

View on phantasyx.com

Hosted MCP Server

npx add-mcp 'https://kempt-horizon-data-toolkit.mcpize.run/mcp'

Installs into Claude Code, Codex, Cursor and more

Documentation

Product	What it does	Get access
Data Cleaner & Validator	Validate emails, format phone numbers, clean CSVs and text	RapidAPI
Page to Markdown & Metadata	Turn any web page or HTML into clean Markdown and metadata	RapidAPI
US Federal Data	SEC EDGAR, Federal Register, USAspending and Grants.gov in one API	RapidAPI
MCP Toolkit	All of the above as 17 tools for AI assistants and agents that support MCP. Listed on MCPize as Kempt Horizon Data Toolkit	MCPize

See all our products at phantasyx.com/apis

Get started

REST APIs on RapidAPI

Open the API's RapidAPI page (links above) and sign in.
Go to Pricing and subscribe to the free Basic plan.
Copy your key from the API's Endpoints tab. It's the X-RapidAPI-Key value.
Send these two headers with every request.

X-RapidAPI-Key: YOUR_RAPIDAPI_KEY
X-RapidAPI-Host: <the API's host, below>

API	Base URL	`X-RapidAPI-Host`
Data Cleaner & Validator	`https://kempt-data-cleaner-validator.p.rapidapi.com`	`kempt-data-cleaner-validator.p.rapidapi.com`
Page to Markdown & Metadata	`https://kempt-page-to-markdown-metadata.p.rapidapi.com`	`kempt-page-to-markdown-metadata.p.rapidapi.com`
US Federal Data	`https://kempt-us-federal-data.p.rapidapi.com`	`kempt-us-federal-data.p.rapidapi.com`

Here's a first call you can copy.

curl "https://kempt-data-cleaner-validator.p.rapidapi.com/v1/email/validate?email=jane@example.com" \
  -H "X-RapidAPI-Key: YOUR_RAPIDAPI_KEY" \
  -H "X-RapidAPI-Host: kempt-data-cleaner-validator.p.rapidapi.com"

A few details apply to every endpoint.

Requests and responses are JSON. Send Content-Type: application/json with POST bodies.
Many endpoints accept both GET (query string) and POST (JSON body) with the same parameters.
List parameters take a JSON array in a POST body, or a comma-separated string in a query string (for example forms=10-K,10-Q).
Dates are always YYYY-MM-DD.

MCP server on MCPize

Open Kempt Horizon Data Toolkit on MCPize and subscribe. The Free plan includes 50 tool calls a month.
Follow MCPize's connection instructions on that page to add the server to any AI assistant or agent that supports remote MCP servers. MCPize gives you the server URL and your personal key.
Then ask your assistant something like "Get Apple's last two years of revenue from SEC filings" or "Turn this page into Markdown," and it picks the matching tool.

The server uses streamable HTTP, and all 17 tools are read-only. The full tool list is further down this page.

Data Cleaner & Validator

Host: kempt-data-cleaner-validator.p.rapidapi.com

This API validates email addresses (syntax, mail exchange (MX) records, disposable domains, role accounts, and typo suggestions), formats phone numbers to the international E.164 standard, cleans messy CSV tables, and normalizes text. We never connect to a recipient's mail server to test whether a mailbox exists.

Validate one email

POST /v1/email/validate, also GET /v1/email/validate?email=...

Parameter	Type	Notes
`email`	string, required	The address to check
`checkMx`	boolean	Default `true`. Looks up MX records over DNS-over-HTTPS

The verdict is one of valid, valid_role, risky or invalid, and the score runs from 0 to 100.

curl -X POST "https://kempt-data-cleaner-validator.p.rapidapi.com/v1/email/validate" \
  -H "X-RapidAPI-Key: YOUR_RAPIDAPI_KEY" \
  -H "X-RapidAPI-Host: kempt-data-cleaner-validator.p.rapidapi.com" \
  -H "Content-Type: application/json" \
  -d '{"email": "Jane.Doe@gmial.com"}'

{
  "input": "Jane.Doe@gmial.com",
  "normalized": "Jane.Doe@gmial.com",
  "local": "Jane.Doe",
  "domain": "gmial.com",
  "syntaxValid": true,
  "reasons": ["possible_typo"],
  "disposable": false,
  "roleAccount": false,
  "freeProvider": false,
  "suggestion": "Jane.Doe@gmail.com",
  "dns": { "checked": true, "domainExists": true, "mxRecords": [], "acceptsMail": true, "fallbackToA": true },
  "verdict": "risky",
  "score": 60,
  "smtpChecked": false
}

Validate up to 100 emails

POST /v1/email/validate-batch

Parameter	Type	Notes
`emails`	string[], required	Up to 100 addresses
`checkMx`	boolean	Default `true`

curl -X POST "https://kempt-data-cleaner-validator.p.rapidapi.com/v1/email/validate-batch" \
  -H "X-RapidAPI-Key: YOUR_RAPIDAPI_KEY" \
  -H "X-RapidAPI-Host: kempt-data-cleaner-validator.p.rapidapi.com" \
  -H "Content-Type: application/json" \
  -d '{"emails": ["a@gmail.com", "b@mailinator.com", "bad@"]}'

{
  "count": 3,
  "summary": { "valid": 1, "risky": 1, "invalid": 1 },
  "results": [
    { "input": "a@gmail.com", "freeProvider": true, "reasons": [], "verdict": "valid", "score": 100 },
    { "input": "b@mailinator.com", "disposable": true, "reasons": ["disposable_domain"], "verdict": "risky" },
    { "input": "bad@", "syntaxValid": false, "verdict": "invalid" }
  ]
}

Each item in results has the same full shape as the single-email response. The items above are trimmed.

Parse a phone number

POST /v1/phone/parse, also GET /v1/phone/parse?phone=...

Parameter	Type	Notes
`phone`	string, required	Any format
`defaultCountry`	string	ISO 3166-1 alpha-2 (for example `US`), used when the number has no `+country` code

curl -X POST "https://kempt-data-cleaner-validator.p.rapidapi.com/v1/phone/parse" \
  -H "X-RapidAPI-Key: YOUR_RAPIDAPI_KEY" \
  -H "X-RapidAPI-Host: kempt-data-cleaner-validator.p.rapidapi.com" \
  -H "Content-Type: application/json" \
  -d '{"phone": "(415) 555-2671", "defaultCountry": "US"}'

{
  "input": "(415) 555-2671",
  "valid": true,
  "possible": true,
  "e164": "+14155552671",
  "national": "(415) 555-2671",
  "international": "+1 415 555 2671",
  "uri": "tel:+14155552671",
  "country": "US",
  "countryCallingCode": "1",
  "nationalNumber": "4155552671",
  "extension": null,
  "type": "FIXED_LINE_OR_MOBILE"
}

Parse up to 100 phone numbers

POST /v1/phone/parse-batch

Parameter	Type	Notes
`phones`	string[], required	Up to 100 numbers
`defaultCountry`	string	Applied to numbers without a country code

curl -X POST "https://kempt-data-cleaner-validator.p.rapidapi.com/v1/phone/parse-batch" \
  -H "X-RapidAPI-Key: YOUR_RAPIDAPI_KEY" \
  -H "X-RapidAPI-Host: kempt-data-cleaner-validator.p.rapidapi.com" \
  -H "Content-Type: application/json" \
  -d '{"phones": ["(415) 555-2671", "+44 20 7946 0958"], "defaultCountry": "US"}'

{
  "count": 2,
  "valid": 2,
  "results": [
    { "input": "(415) 555-2671", "valid": true, "e164": "+14155552671", "country": "US", "type": "FIXED_LINE_OR_MOBILE" },
    { "input": "+44 20 7946 0958", "valid": true, "e164": "+442079460958", "national": "020 7946 0958", "country": "GB", "type": "FIXED_LINE" }
  ]
}

Clean a CSV or JSON table

POST /v1/csv/clean

Send csv (raw CSV or TSV text, up to 2 MB and 5,000 rows, delimiter detected automatically) or rows (an array of arrays with a header row, or an array of objects). You don't have to set any of the options below.

Option	Default	Notes
`delimiter`	auto	Force `,` `;` `\
`normalizeHeaders`	`snake`	`snake`, `camel`, `kebab` or `none`
`trim`, `dedupe`, `dropEmptyRows`	`true`	Trim cells, remove duplicate rows, drop empty rows
`dedupeColumns`	all	Deduplicate on these columns only (case-insensitive)
`dropEmptyColumns`	`false`
`emailColumns`, `dateColumns`	auto	Column list, or `"none"` to skip
`phoneColumns`	none	Columns to convert to E.164
`defaultCountry`	`US`	For phone columns
`dateOrder`	`MDY`	`MDY` or `DMY` for ambiguous dates
`output`	`both`	`json`, `csv` or `both`
`outputDelimiter`	`,`

curl -X POST "https://kempt-data-cleaner-validator.p.rapidapi.com/v1/csv/clean" \
  -H "X-RapidAPI-Key: YOUR_RAPIDAPI_KEY" \
  -H "X-RapidAPI-Host: kempt-data-cleaner-validator.p.rapidapi.com" \
  -H "Content-Type: application/json" \
  -d '{"csv": "First Name,E-mail,Signup Date\n Jane ,JANE@Example.com,10/21/2026\nJane,jane@example.com,10/21/2026\n,,\n", "options": {"output": "json"}}'

{
  "delimiter": ",",
  "headers": ["first_name", "email", "signup_date"],
  "report": {
    "rowsIn": 3, "rowsOut": 1, "emptyRowsRemoved": 1, "duplicatesRemoved": 1,
    "cellsTrimmed": 1, "emailsNormalized": 1, "invalidEmails": 0,
    "phonesNormalized": 0, "invalidPhones": 0, "datesNormalized": 2,
    "unparsedDates": 0, "raggedRowsFixed": 0, "columnsRemoved": []
  },
  "columns": [
    { "name": "first_name", "type": "string", "empty": 0, "distinct": 1 },
    { "name": "email", "type": "email", "empty": 0, "distinct": 1 },
    { "name": "signup_date", "type": "date", "empty": 0, "distinct": 1 }
  ],
  "rows": [
    { "first_name": "Jane", "email": "jane@example.com", "signup_date": "2026-10-21" }
  ]
}

Clean and normalize text

POST /v1/text/clean

Parameter	Type	Notes
`text`	string, required
`options.stripHtml`	boolean	Default `false`
`options.decodeEntities`	boolean	Default `true`
`options.unicode`	string	`NFC` (default) or `NFKC`
`options.removeControlChars`	boolean	Default `true`
`options.asciiPunctuation`	boolean	Smart quotes and dashes to ASCII. Default `false`
`options.removeEmoji`, `removeUrls`, `removeEmails`	boolean	Default `false`
`options.collapseWhitespace`	boolean	Default `true`
`options.case`	string	`lower`, `upper`, `sentence` or `title`
`options.maxLength`	integer	Cut the output to this length

curl -X POST "https://kempt-data-cleaner-validator.p.rapidapi.com/v1/text/clean" \
  -H "X-RapidAPI-Key: YOUR_RAPIDAPI_KEY" \
  -H "X-RapidAPI-Host: kempt-data-cleaner-validator.p.rapidapi.com" \
  -H "Content-Type: application/json" \
  -d '{"text": "<p>Hello&nbsp;&nbsp;“world” 👋</p>", "options": {"stripHtml": true, "asciiPunctuation": true, "removeEmoji": true}}'

{
  "text": "Hello \"world\"",
  "stats": { "charactersIn": 34, "charactersOut": 13, "words": 2, "lines": 1 }
}

Page to Markdown & Metadata

Host: kempt-page-to-markdown-metadata.p.rapidapi.com

This API turns a public web page into clean Markdown and structured metadata. Send a url and we fetch the page for you, or send the html you already have.

All three endpoints accept the same parameters by POST (JSON body) or GET (query string, where url is required).

Parameter	Type	Notes
`url`	string	Public http(s) URL. Optional when you send `html`, and then used as the base for links
`html`	string	Raw HTML to process instead of fetching (max 3 MB). POST only
`mainContentOnly`	boolean	Default `true`. Removes nav, headers, footers, sidebars, cookie banners and ads
`maxChars`	integer	Default `100000` (range 1,000 to 500,000). Caps the Markdown or text length
`include`	string or string[]	Any of `metadata`, `markdown`, `text`, `links`, `images`, `headings`

When we fetch a page, we honor its robots.txt, stop after 10 seconds, and cap the download at 3 MB. We read the HTML the server sends and don't run a browser, so content that only appears after scripts run won't be included. We block private, local, and raw IP addresses. If a site's robots.txt disallows a page, you can send its HTML in html instead.

Everything in one call

POST /v1/extract, also GET /v1/extract?url=...

By default this returns metadata, markdown, links and headings.

curl -X POST "https://kempt-page-to-markdown-metadata.p.rapidapi.com/v1/extract" \
  -H "X-RapidAPI-Key: YOUR_RAPIDAPI_KEY" \
  -H "X-RapidAPI-Host: kempt-page-to-markdown-metadata.p.rapidapi.com" \
  -H "Content-Type: application/json" \
  -d '{"url": "https://example.com/", "include": ["metadata", "markdown"]}'

{
  "source": "url",
  "finalUrl": "https://example.com/",
  "httpStatus": 200,
  "contentType": "text/html; charset=utf-8",
  "redirects": [],
  "htmlTruncated": false,
  "metadata": {
    "url": "https://example.com/",
    "title": "Example Domain",
    "language": "en",
    "favicon": "https://example.com/favicon.ico",
    "openGraph": {},
    "twitter": {},
    "jsonLd": []
  },
  "stats": { "htmlBytes": 577, "words": 25, "readingTimeMinutes": 1, "mainContentOnly": true },
  "truncated": false,
  "markdown": "This domain is for use in documentation examples without needing permission. ..."
}

HTML or URL to Markdown

POST /v1/markdown, also GET /v1/markdown?url=...

Returns the main content as Markdown with absolute links, fenced code blocks and tables. By default it returns markdown only.

curl -X POST "https://kempt-page-to-markdown-metadata.p.rapidapi.com/v1/markdown" \
  -H "X-RapidAPI-Key: YOUR_RAPIDAPI_KEY" \
  -H "X-RapidAPI-Host: kempt-page-to-markdown-metadata.p.rapidapi.com" \
  -H "Content-Type: application/json" \
  -d '{"html": "<article><h1>Hello</h1><p>World <a href=\"/x\">link</a></p></article>", "url": "https://example.com/"}'

{
  "source": "html",
  "stats": { "htmlBytes": 67, "words": 3, "readingTimeMinutes": 1, "mainContentOnly": true },
  "truncated": false,
  "markdown": "# Hello\n\nWorld  [link](https://example.com/x)"
}

Link preview and SEO metadata

POST /v1/metadata, also GET /v1/metadata?url=...

This returns the title, description, canonical URL, language, Open Graph, Twitter Card, JSON-LD, favicon and icons, RSS and Atom feeds, and hreflang alternates. It only reads the page's <head>, so it skips the rest of the page.

curl "https://kempt-page-to-markdown-metadata.p.rapidapi.com/v1/metadata?url=https%3A%2F%2Fexample.com%2F" \
  -H "X-RapidAPI-Key: YOUR_RAPIDAPI_KEY" \
  -H "X-RapidAPI-Host: kempt-page-to-markdown-metadata.p.rapidapi.com"

{
  "source": "url",
  "finalUrl": "https://example.com/",
  "httpStatus": 200,
  "metadata": {
    "url": "https://example.com/",
    "canonicalUrl": null,
    "title": "Example Domain",
    "description": null,
    "language": "en",
    "siteName": null,
    "author": null,
    "publishedTime": null,
    "image": null,
    "favicon": "https://example.com/favicon.ico",
    "icons": [],
    "openGraph": {},
    "twitter": {},
    "jsonLdTypes": [],
    "jsonLd": [],
    "feeds": [],
    "alternateLanguages": []
  },
  "stats": { "htmlBytes": 577 }
}

US Federal Data

Host: kempt-us-federal-data.p.rapidapi.com

This API brings EDGAR from the Securities and Exchange Commission (SEC), the Federal Register, USAspending.gov, and Grants.gov into one place, with consistent camelCase JSON and YYYY-MM-DD dates. The data comes only from official government APIs that need no key, it's in the public domain, and we follow the SEC's fair-access rules. We aren't affiliated with any government agency.

For company endpoints, pass either ticker (for example AAPL) or cik, the SEC's Central Index Key (for example 320193).

Search companies

GET /v1/sec/companies

Parameter	Type	Notes
`q`	string, required	Company name or ticker fragment (2+ characters)
`limit`	integer	Default 10, max 50

curl "https://kempt-us-federal-data.p.rapidapi.com/v1/sec/companies?q=apple&limit=3" \
  -H "X-RapidAPI-Key: YOUR_RAPIDAPI_KEY" \
  -H "X-RapidAPI-Host: kempt-us-federal-data.p.rapidapi.com"

{
  "query": "apple",
  "count": 3,
  "results": [
    { "cik": "0000320193", "ticker": "AAPL", "name": "Apple Inc." },
    { "cik": "0001134982", "ticker": "AAPI", "name": "Apple iSports Group, Inc." },
    { "cik": "0001418121", "ticker": "APLE", "name": "Apple Hospitality REIT, Inc." }
  ]
}

Company profile

GET /v1/sec/company

Parameter	Type	Notes
`ticker` or `cik`	string	One of the two is required

curl "https://kempt-us-federal-data.p.rapidapi.com/v1/sec/company?ticker=AAPL" \
  -H "X-RapidAPI-Key: YOUR_RAPIDAPI_KEY" \
  -H "X-RapidAPI-Host: kempt-us-federal-data.p.rapidapi.com"

{
  "cik": "0000320193",
  "name": "Apple Inc.",
  "tickers": ["AAPL"],
  "exchanges": ["Nasdaq"],
  "entityType": "operating",
  "sic": "3571",
  "sicDescription": "Electronic Computers",
  "category": "Large accelerated filer",
  "fiscalYearEnd": "0926",
  "stateOfIncorporation": "CA",
  "businessAddress": { "street1": "ONE APPLE PARK WAY", "city": "CUPERTINO", "state": "CA", "zip": "95014" },
  "formerNames": [{ "name": "APPLE COMPUTER INC", "from": "1994-01-26", "to": "2007-01-04" }],
  "latestFiling": { "form": "4", "filingDate": "2026-10-19", "accessionNumber": "0001140361-26-040127" },
  "edgarUrl": "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000320193",
  "source": "SEC EDGAR submissions API"
}

Recent filings

GET /v1/sec/filings

Parameter	Type	Notes
`ticker` or `cik`	string	One of the two is required
`forms`	string	Comma-separated form types, for example `10-K,10-Q,8-K`
`dateFrom`, `dateTo`	date	`YYYY-MM-DD`
`limit`	integer	Default 20, max 200

curl "https://kempt-us-federal-data.p.rapidapi.com/v1/sec/filings?ticker=AAPL&forms=10-K&limit=2" \
  -H "X-RapidAPI-Key: YOUR_RAPIDAPI_KEY" \
  -H "X-RapidAPI-Host: kempt-us-federal-data.p.rapidapi.com"

{
  "cik": "0000320193",
  "name": "Apple Inc.",
  "count": 2,
  "filings": [
    {
      "accessionNumber": "0000320193-25-000079",
      "form": "10-K",
      "filingDate": "2025-10-31",
      "reportDate": "2025-09-27",
      "items": [],
      "primaryDocument": "aapl-20250927.htm",
      "isXBRL": true,
      "primaryDocumentUrl": "https://www.sec.gov/Archives/edgar/data/320193/000032019325000079/aapl-20250927.htm",
      "filingIndexUrl": "https://www.sec.gov/Archives/edgar/data/320193/000032019325000079/0000320193-25-000079-index.htm"
    }
  ],
  "note": "Covers the most recent ~1,000 filings (EDGAR \"recent\" set)."
}

Key financials (XBRL)

GET /v1/sec/financials

Parameter	Type	Notes
`ticker` or `cik`	string	One of the two is required
`period`	string	`annual` (default) or `quarterly`
`metrics`	string	Comma-separated subset of `revenue`, `netIncome`, `operatingIncome`, `grossProfit`, `epsDiluted`, `epsBasic`, `totalAssets`, `totalLiabilities`, `stockholdersEquity`, `cash`, `operatingCashFlow`, `sharesOutstanding`
`limit`	integer	Periods per metric. Default 5, max 20

Values come straight from the filings' eXtensible Business Reporting Language (XBRL) data and aren't adjusted. If a company doesn't use the usual XBRL tag for a metric, we use the closest one it does report.

curl "https://kempt-us-federal-data.p.rapidapi.com/v1/sec/financials?ticker=AAPL&metrics=revenue,netIncome&limit=2" \
  -H "X-RapidAPI-Key: YOUR_RAPIDAPI_KEY" \
  -H "X-RapidAPI-Host: kempt-us-federal-data.p.rapidapi.com"

{
  "cik": "0000320193",
  "ticker": "AAPL",
  "period": "annual",
  "metrics": {
    "revenue": {
      "concept": "us-gaap:RevenueFromContractWithCustomerExcludingAssessedTax",
      "label": "Revenue from Contract with Customer, Excluding Assessed Tax",
      "unit": "USD",
      "values": [
        {
          "periodStart": "2024-09-29",
          "periodEnd": "2025-09-27",
          "value": 416161000000,
          "fiscalYear": 2025,
          "fiscalPeriod": "FY",
          "form": "10-K",
          "filed": "2025-10-31",
          "accessionNumber": "0000320193-25-000079"
        }
      ]
    },
    "netIncome": { "concept": "us-gaap:NetIncomeLoss", "unit": "USD", "values": [{ "periodEnd": "2025-09-27", "value": 112010000000 }] }
  },
  "source": "SEC XBRL companyconcept API (as reported; not adjusted)"
}

Full-text filing search

GET /v1/sec/search

Parameter	Type	Notes
`q`	string	EDGAR full-text query. Put exact phrases in quotes. You must send `q`, `forms`, or both
`forms`	string	Comma-separated form types
`ticker` or `cik`	string	Optional company filter
`dateFrom`, `dateTo`	date	`YYYY-MM-DD` (coverage starts in 2001)
`limit`	integer	Default 20, max 100

curl "https://kempt-us-federal-data.p.rapidapi.com/v1/sec/search?q=%22artificial%20intelligence%22&forms=10-K&limit=2" \
  -H "X-RapidAPI-Key: YOUR_RAPIDAPI_KEY" \
  -H "X-RapidAPI-Host: kempt-us-federal-data.p.rapidapi.com"

{
  "query": "\"artificial intelligence\"",
  "forms": ["10-K"],
  "totalDocumentHits": 10000,
  "count": 2,
  "filings": [
    {
      "accessionNumber": "0001161697-21-000289",
      "form": "10-K",
      "fileDate": "2021-06-01",
      "periodEnding": "2021-02-28",
      "cik": "0001498148",
      "companyName": "Artificial Intelligence Technology Solutions Inc.",
      "ticker": "AITX",
      "primaryDocumentUrl": "https://www.sec.gov/Archives/edgar/data/1498148/000116169721000289/form_10-k.htm",
      "matchedDocuments": [
        { "file": "form_10-k.htm", "type": "10-K", "description": "FORM 10-K ANNUAL REPORT FOR 02-28-2021", "url": "https://www.sec.gov/Archives/edgar/data/1498148/000116169721000289/form_10-k.htm" }
      ]
    }
  ]
}

Federal Register documents

GET /v1/federal-register/documents

Parameter	Type	Notes
`keywords`	string	Comma-separated search terms (matches any)
`types`	string	`RULE`, `PRORULE`, `NOTICE`, `PRESDOCU`
`agencies`	string	Agency names or slugs, for example `environmental-protection-agency`
`lastDays` or `dateFrom` / `dateTo`	integer / date	Publication window
`significantOnly`, `openForCommentOnly`	boolean
`docketId`, `rin`	string
`cfrTitle`, `cfrPart`	integer
`order`	string	`newest`, `oldest` or `relevance`
`limit`	integer	Default 20, max 100

curl "https://kempt-us-federal-data.p.rapidapi.com/v1/federal-register/documents?keywords=artificial%20intelligence&types=RULE&limit=2" \
  -H "X-RapidAPI-Key: YOUR_RAPIDAPI_KEY" \
  -H "X-RapidAPI-Host: kempt-us-federal-data.p.rapidapi.com"

{
  "total": 186,
  "count": 2,
  "documents": [
    {
      "documentNumber": "2026-20890",
      "title": "Project-Level Predecisional Administrative Review Process",
      "type": "Rule",
      "action": "Final rule.",
      "agencies": ["Agriculture Department", "Forest Service"],
      "publicationDate": "2026-10-13",
      "effectiveDate": "2026-11-12",
      "commentsCloseDate": null,
      "regulationIdNumbers": ["0596-AD69"],
      "cfrReferences": [{ "title": 36, "part": "218", "chapter": null }],
      "citation": "91 FR 64760",
      "pdfUrl": "https://www.govinfo.gov/content/pkg/FR-2026-10-13/pdf/2026-20890.pdf",
      "commentUrl": null
    }
  ]
}

Federal awards (USAspending)

POST /v1/usaspending/awards

Parameter	Type	Notes
`category`	string	`contracts` (default), `idvs`, `grants`, `loans`, `direct_payments` or `other`
`keywords`	string[]	Each keyword must be 3+ characters
`recipients`	string[]	Recipient names
`agencies`	string[]	Awarding top-tier agency names, for example `Department of Defense`
`naics`, `psc`	string[]	Industry and product codes (contracts)
`assistanceListings`	string[]	ALN/CFDA numbers (assistance)
`states`	string[]	Place-of-performance state codes
`setAsides`	string[]	Set-aside type codes
`minAmount`, `maxAmount`	number
`lastDays` or `dateFrom` / `dateTo`	integer / date	Defaults to the last 90 days
`sortBy`	string	`lastModified` or `amount`
`page`, `limit`	integer	Page default 1. Limit default 20, max 100

curl -X POST "https://kempt-us-federal-data.p.rapidapi.com/v1/usaspending/awards" \
  -H "X-RapidAPI-Key: YOUR_RAPIDAPI_KEY" \
  -H "X-RapidAPI-Host: kempt-us-federal-data.p.rapidapi.com" \
  -H "Content-Type: application/json" \
  -d '{"category": "contracts", "keywords": ["cybersecurity"], "lastDays": 30, "limit": 2}'

{
  "category": "contracts",
  "timePeriod": ["2026-09-22", "2026-10-22"],
  "sortBy": "Last Modified Date",
  "page": 1,
  "hasNext": true,
  "count": 2,
  "awards": [
    {
      "awardId": "89303026CMA000146",
      "category": "contract",
      "awardType": "DEFINITIVE CONTRACT",
      "recipientName": "AXIS MANAGEMENT GROUP, LLC",
      "amount": 200000,
      "startDate": "2026-09-24",
      "endDate": "2027-03-31",
      "awardingAgency": "Department of Energy",
      "naics": { "code": "519290", "description": "WEB SEARCH PORTALS AND ALL OTHER INFORMATION SERVICES" },
      "url": "https://www.usaspending.gov/award/CONT_AWD_89303026CMA000146_8900_-NONE-_-NONE-"
    }
  ]
}

The response above is trimmed to one award. Full awards also include recipientUei, description, awardingSubAgency, fundingAgency, psc, placeOfPerformance and recipientUrl.

Search grant opportunities

POST /v1/grants/search

Parameter	Type	Notes
`keywords`	string[]	Matches any
`statuses`	string[]	`posted`, `forecasted`, `closed`, `archived`. Default `posted,forecasted`
`agencies`	string[]	Agency codes, for example `HHS-NIH11`
`fundingCategories`	string[]	Category codes, for example `ED`, `HL`, `ST`
`eligibilities`	string[]	Eligibility codes
`aln`	string	Assistance listing number
`includeDetails`	boolean	Adds award ceiling and floor, description, eligibility and contacts for the first 10 results. Default `false`
`limit`	integer	Default 20, max 50

curl -X POST "https://kempt-us-federal-data.p.rapidapi.com/v1/grants/search" \
  -H "X-RapidAPI-Key: YOUR_RAPIDAPI_KEY" \
  -H "X-RapidAPI-Host: kempt-us-federal-data.p.rapidapi.com" \
  -H "Content-Type: application/json" \
  -d '{"keywords": ["climate resilience"], "limit": 10}'

{
  "total": 3,
  "count": 3,
  "grants": [
    {
      "opportunityId": "347414",
      "opportunityNumber": "EDA-DISASTER-2023",
      "title": "FY 2023 Disaster Supplemental",
      "agencyCode": "DOC-EDA",
      "agency": "Economic Development Administration",
      "status": "posted",
      "docType": "synopsis",
      "postedDate": "2023-04-24",
      "closeDate": null,
      "alnNumbers": ["11.307"],
      "url": "https://www.grants.gov/search-results-detail/347414"
    }
  ],
  "source": "Grants.gov search2 API"
}

Grant opportunity details

GET /v1/grants/opportunity

Parameter	Type	Notes
`opportunityId`	string, required	Grants.gov opportunity ID (from `/v1/grants/search`)

curl "https://kempt-us-federal-data.p.rapidapi.com/v1/grants/opportunity?opportunityId=350338" \
  -H "X-RapidAPI-Key: YOUR_RAPIDAPI_KEY" \
  -H "X-RapidAPI-Host: kempt-us-federal-data.p.rapidapi.com"

{
  "opportunityId": "350338",
  "opportunityNumber": "2023-NIST-CHIPS-SMME-01",
  "title": "CHIPS Incentives Program &ndash; Facilities for Semiconductor Materials and Manufacturing Equipment",
  "agency": "DOC-NIST",
  "category": "Discretionary",
  "description": "The CHIPS Incentives Program aims to catalyze long-term economically sustainable growth in the domestic semiconductor industry ...",
  "awardCeiling": null,
  "awardFloor": null,
  "costSharing": false,
  "fundingInstruments": ["Cooperative Agreement"],
  "fundingCategories": ["Science and Technology and other Research and Development"],
  "closeDate": null,
  "lastUpdated": "2025-10-16",
  "moreInfoUrl": "https://applications.chips.gov/s/",
  "aln": [{ "number": "11.037", "title": "CHIPS Incentives Program" }],
  "documents": [
    { "name": "2023-NIST-CHIPS-SMME-01-Amendment.pdf", "url": "https://apply07.grants.gov/grantsws/rest/opportunity/att/download/349581" }
  ]
}

The full response also includes eligibleApplicants, eligibilityNotes, estimatedTotalFunding, expectedNumberOfAwards, responseDateNote and agencyContact.

MCP tools

Our MCP Toolkit (listed on MCPize as Kempt Horizon Data Toolkit) gives your assistant all 17 tools from one remote server. Each tool takes the same parameters as the REST endpoint it matches.

Read the web

Tool	What it does
`page_to_markdown`	Fetch a public page (robots.txt honored) or take raw HTML and return the main content as clean Markdown with absolute links
`page_metadata`	Title, description, canonical URL, Open Graph and Twitter tags, JSON-LD, favicon and feeds for a URL or HTML
`page_extract`	Markdown, metadata, links and headings in one call. Use `include` to pick parts

Clean data

Tool	What it does
`validate_email`	Check one email for syntax, MX records, disposable, role and free-provider flags, and typos, then return a verdict and score
`validate_emails`	Validate up to 100 email addresses at once
`parse_phone`	Validate a phone number and return E.164, national and international formats, country and line type
`clean_csv`	Normalize headers, trim, dedupe, drop empty rows, fix emails, phones and dates, and infer column types
`clean_text`	Strip HTML, decode entities, normalize Unicode, whitespace and punctuation, remove emoji, URLs or emails, change case

US government data

Tool	What it does
`sec_company_search`	Find public companies by name or ticker and get their CIK
`sec_company_profile`	Get a company's Standard Industrial Classification (SIC) industry, state of incorporation, fiscal year end, addresses, exchanges, and former names
`sec_company_filings`	Recent SEC filings (10-K, 10-Q, 8-K and more) with document URLs
`sec_company_financials`	Get revenue, net income, earnings per share (EPS), assets, cash, and more from XBRL data, annual or quarterly
`sec_filing_search`	Full-text search across EDGAR filings, filtered by form and date
`federal_register_search`	Search federal rules, proposed rules and notices by agency, type, date or open comment period
`usaspending_awards`	Search federal contract and grant awards to see who got paid, how much, and by which agency
`grants_search`	Find open and forecasted federal grant opportunities by keyword, agency or category
`grant_details`	Full details for one Grants.gov opportunity

Plans and rate limits

All quotas are hard limits with no overage charges. When you reach your monthly quota, requests stop until the next cycle or until you upgrade. You can change plans at any time on RapidAPI or MCPize.

RapidAPI plans

Plan	Price	Data Cleaner & Validator	Page to Markdown & Metadata	US Federal Data
Basic	$0	50 req/mo, 1 req/s	50 req/mo, 1 req/s	50 req/mo, 1 req/s
Pro	$9.99/mo	10,000 req/mo	5,000 req/mo	5,000 req/mo
Ultra	$29.99/mo	50,000 req/mo	25,000 req/mo	25,000 req/mo
Mega	$79.99/mo	200,000 req/mo	100,000 req/mo	100,000 req/mo

Batch endpoints count as one request, so up to 100 emails or phone numbers cost a single call.

MCPize plans (Kempt Horizon Data Toolkit)

Plan	Price	Included tool calls
Free	$0	50 / month
Pro	$9.99/mo	5,000 / month
Ultra	$29.99/mo	25,000 / month

Size limits

Limit	Value
Request body	2 MB
Batch email or phone	100 items
CSV	2 MB, 5,000 rows
Fetched page or `html`	3 MB, 10-second timeout
Markdown or text output	`maxChars`, 1,000 to 500,000 (default 100,000)

Errors

Errors come back as JSON with a matching HTTP status, like this.

{ "error": { "code": "missing_parameter", "message": "\"email\" is required." } }

Some errors add extra fields, such as retryAfterSeconds on rate_limited, upstreamStatus on upstream_status, or upstreamDetail when a government source rejects a filter.

HTTP	Code	Meaning
400	`missing_parameter`	A required parameter is missing, for example `email`, `url` or `html`, `ticker` or `cik`, or `opportunityId`
400	`missing_input`	`/v1/csv/clean` needs `csv` or `rows`
400	`invalid_json` / `invalid_body`	The body isn't valid JSON or isn't a JSON object
400	`batch_too_large`	More than 100 items in a batch
400	`invalid_url` / `blocked_url`	The URL isn't a public http(s) URL on port 80 or 443, or it points to a private or local address
400	`missing_html`	There was no HTML to process
400	`invalid_include`	Unknown `include` value
400	`invalid_date`	Dates must be `YYYY-MM-DD`
400	`invalid_cik` / `invalid_metric` / `invalid_category` / `invalid_keyword`	Bad value for that parameter. USAspending keywords need 3+ characters
400	`invalid_filters`	The Federal Register or USAspending rejected the filter combination
401	`unauthorized`	The request didn't come through RapidAPI or MCPize with a valid key
403	`robots_disallowed`	The site's robots.txt doesn't allow fetching that page. Send the HTML in `html` instead
404	`not_found` / `ticker_not_found`	Unknown endpoint, unknown ticker, or record not found
405	`method_not_allowed`	Wrong HTTP method for this path
413	`payload_too_large` / `too_many_rows`	Body over 2 MB, or CSV over 5,000 rows
415	`unsupported_content_type`	The fetched URL isn't HTML
429	`rate_limited`	Too many requests in a short time. Wait `retryAfterSeconds` and try again
502	`fetch_failed` / `upstream_status`	The page couldn't be fetched (timeout or network error) or returned an error status
502	`upstream_error` / `upstream_unreachable`	A government data source failed or didn't respond. Try again shortly
503	`upstream_busy`	A government data source is rate limiting. Try again in a few seconds
508	`too_many_redirects`	The page redirected too many times
500	`internal_error`	Something unexpected went wrong on our side. Try again, and contact us if it keeps happening

RapidAPI and MCPize may also return their own errors, for example for a missing or invalid key or a used-up monthly quota.

In MCP clients, a failed tool call returns isError: true with a message such as missing_parameter: "q" (2+ characters) is required.

Privacy and fair use

We don't store or log the emails, phone numbers, tables, text, or HTML you send, and no API keeps anything between requests.
When we fetch a page for you, we check the site's robots.txt first and follow it, and we block private and local addresses.
Federal data comes only from official public government APIs, and we follow the SEC's fair-access rules.
Every MCP tool is read-only and never changes anything on your behalf.

Support

If you have a question, find a bug, or want a feature, contact us.