Guide · US Company Filings and Financials

10-K and XBRL company facts by API: limits and fields

By Tan Solutions · Published · Updated

Short answer

The SEC publishes every public company’s filing history and the financial facts tagged in its 10-K and 10-Q reports through a free JSON API at data.sec.gov. You need no account and no API key. You do need to say who you are in the User-Agent header, and you must stay under the current maximum of 10 requests per second (SEC, Accessing EDGAR Data, read 15 November 2026).

1. What the public filings API offers

The SEC’s API page says data.sec.gov “was created to host RESTful data APIs delivering JSON-formatted data” and that these APIs “do not require any authentication or API keys” (SEC, EDGAR APIs, last updated 8 April 2025, read 15 November 2026). Two kinds of data are in it:

  • Submissions: the filing history of each filer.
  • XBRL data: the financial statement facts from forms 10-Q, 10-K, 8-K, 20-F, 40-F, 6-K and their variants.

The JSON is updated throughout the day as filings come in. The SEC gives a typical delay of under a second for the submissions API and under a minute for the XBRL APIs, longer at peak filing times (same page).

There are four endpoints you will use:

Endpoint What it returns
https://data.sec.gov/submissions/CIK##########.json Company name, former names, tickers, exchanges and recent filings
https://data.sec.gov/api/xbrl/companyfacts/CIK##########.json Every XBRL fact the company has reported, in one file
https://data.sec.gov/api/xbrl/companyconcept/CIK##########/us-gaap/<Tag>.json Every value of one fact (one tag) for one company
https://data.sec.gov/api/xbrl/frames/us-gaap/<Tag>/USD/CY2019Q1I.json One value of one fact for every company, for one period

The ########## is the company’s CIK (Central Index Key), padded to 10 digits with leading zeros. Apple’s CIK is 320193, so its submissions file is CIK0000320193.json.

2. Company facts vs filings

The two halves of the API answer different questions.

Submissions answer “what did this company file, and when?” The file holds at least one year of filings or the 1,000 most recent, whichever is more. Older filings sit in extra JSON files that the main file points to (SEC API page). Each filing has a form type, a filing date, a report date and an accession number you can turn into a link to the document.

Company facts answer “what number did this company report?” The SEC aggregates only facts that use a standard taxonomy (us-gaap, ifrs-full, dei or srt) and that apply to the whole company, so they are comparable across companies and across years (SEC API page). Custom tags a company invents for itself are left out.

If you want one fact for one company, the company concept endpoint is the smallest call. Here is the start of the real response for Apple’s net income (read 15 November 2026):

{
  "cik": 320193,
  "taxonomy": "us-gaap",
  "tag": "NetIncomeLoss",
  "label": "Net Income (Loss) Attributable to Parent",
  "entityName": "Apple Inc.",
  "units": {
    "USD": [
      { "start": "2006-10-01", "end": "2007-09-29", "val": 3496000000,
        "accn": "0001193125-09-214859", "fy": 2009, "fp": "FY",
        "form": "10-K", "filed": "2009-10-27" }
    ]
  }
}

Note what this shows: the same period appears again and again, once for every later filing that repeated it, and amendments (10-K/A) appear too. fy and fp describe the filing the value came from, not always the period the value covers. Use start, end and form to pick the value you want.

3. The user-agent rule and the rate limit

These two rules decide whether your script works at all.

Rate limit. The SEC states: “Current max request rate: 10 requests/second”, and it “reserves the right to limit request rates to preserve fair access for all users” (Accessing EDGAR Data, read 15 November 2026). The Webmaster FAQ repeats the 10 per second figure and says it “is carefully monitored” (SEC Webmaster FAQ, Developers, read 15 November 2026).

User agent. The SEC asks you to “declare your user agent in request headers” and gives this sample (same pages):

User-Agent: Sample Company Name AdminContact@<sample company domain>.com
Accept-Encoding: gzip, deflate
Host: www.sec.gov

For data.sec.gov the Host is data.sec.gov. The FAQ lists an “Undeclared Automated Tool” error as the message you get when the user agent is missing.

Two more rules from the same pages: data.sec.gov does not support CORS, so you cannot call it from JavaScript in a web page; and the SEC does not give technical support for scripted downloads.

A minimal Python call:

import requests

headers = {"User-Agent": "Your Company you@your-company.com"}
url = "https://data.sec.gov/api/xbrl/companyconcept/CIK0000320193/us-gaap/NetIncomeLoss.json"
data = requests.get(url, headers=headers, timeout=30).json()
annual = [f for f in data["units"]["USD"] if f["form"] == "10-K"]
print(annual[-1])

4. Fields that matter: revenue, net income, shares

Companies do not all use the same tag for the same idea. A few common tags:

  • Revenue: us-gaap:Revenues or us-gaap:RevenueFromContractWithCustomerExcludingAssessedTax; older filings may use us-gaap:SalesRevenueNet; IFRS filers use ifrs-full:Revenue.
  • Net income: us-gaap:NetIncomeLoss (label “Net Income (Loss) Attributable to Parent”, as in the response above), or us-gaap:ProfitLoss.
  • Shares outstanding: dei:EntityCommonStockSharesOutstanding, from the cover page of the filing.

So “revenue” for a list of companies means trying more than one tag per company and keeping the latest annual value. Missing values are normal: a fact exists only if the company tagged it.

The frames endpoint is the other way round: one tag, one period, every company. Its period codes are CY2019 for a year, CY2019Q1 for a quarter and CY2019Q1I for a point in time. The SEC warns that frames are fitted to calendar periods, so a company with a September year end is matched to the closest calendar year (SEC API page).

5. Doing it for 500 tickers

By hand, the job has four steps.

  1. Map tickers to CIKs. The SEC publishes company_tickers.json with ticker, CIK and company name, and says it updates the file periodically “but do[es] not guarantee accuracy or scope” (Webmaster FAQ).
  2. Fetch two files per company: submissions and company facts. That is 1,000 requests for 500 tickers.
  3. Pace yourself under 10 per second, with your user agent on every request, and back off if you are refused.
  4. Pick the facts: for each metric, try the tag list, keep annual 10-K values, take the latest period, and handle amendments.

If you need every company rather than a list, the SEC’s own answer is the bulk files: companyfacts.zip and submissions.zip, “recompiled nightly” (the API page puts the update at about 3:00 a.m. ET). The SEC calls them “the most efficient means to fetch large amounts of API data”.

With our tool. US Company Filings and Financials does steps 1 to 4 for up to 1,000 tickers or CIK numbers per run. It reads the official ticker list, sends your name and email in the user-agent header (they go only to the SEC and never into your results), and paces itself at 2 requests per second by default and never more than 5. A real input:

{
  "companies": ["AAPL", "MSFT"],
  "forms": ["10-K", "10-Q"],
  "companyName": "Your Company",
  "contactEmail": "you@your-company.com",
  "maxFilingsPerCompany": 20,
  "includeFacts": true
}

And a real output row from a run on 13 November 2026, shortened to one filing and two of the twelve facts:

{
  "input": "AAPL",
  "status": "ok",
  "cik": "0000320193",
  "name": "Apple Inc.",
  "tickers": ["AAPL"],
  "exchanges": ["Nasdaq"],
  "sicDescription": "Electronic Computers",
  "fiscalYearEnd": "0926",
  "filingsMatched": 315,
  "filingsReturned": 20,
  "filings": [{ "form": "10-K", "filingDate": "2026-11-03", "reportDate": "2026-09-26",
    "filingUrl": "https://www.sec.gov/Archives/edgar/data/320193/000032019326000026/aapl-20260926.htm" }],
  "keyFacts": [
    { "metric": "revenue", "unit": "USD", "value": 478834000000, "fiscalYear": 2026, "form": "10-K", "periodEnd": "2026-09-26" },
    { "metric": "netIncome", "unit": "USD", "value": 132175000000, "fiscalYear": 2026, "form": "10-K", "periodEnd": "2026-09-26" }
  ],
  "scrapedAt": "2026-11-13T09:40:19.578Z"
}

The price is from USD 0.50 per 1,000 companies (USD 0.80 on the Apify free plan), plus Apify’s own charge for starting a run. Tickers that are not found, invalid entries and companies it could not fetch are free. In our test, 100 companies took about 4 minutes. Run it on the Apify Store.

6. Limits to know

  • Only tagged facts. The XBRL APIs return what companies tagged with standard taxonomies; custom tags are not aggregated (SEC API page).
  • Filings are linked, not parsed. The API gives facts and filing metadata. The text of a 10-K is in the filing document itself.
  • The ticker map can lag. Tickers change after mergers and renames; the CIK never changes.
  • No browser calls. No CORS on data.sec.gov.
  • Respect the limit. The SEC can limit request rates for anyone who does not.
  • Our tool: at most 1,000 companies per run, no “all companies” mode (use the SEC bulk files for that), no full filing text, and insiders and other individual filers are skipped unless you turn them on. Not affiliated with, endorsed or approved by the U.S. Securities and Exchange Commission.

Sources

All read on 15 November 2026.

  1. U.S. Securities and Exchange Commission, EDGAR Application Programming Interfaces (APIs), last reviewed or updated 8 April 2025.
  2. U.S. Securities and Exchange Commission, Accessing EDGAR Data, page dated 23 March 2021, current rate stated on the page when read.
  3. U.S. Securities and Exchange Commission, Webmaster Frequently Asked Questions, Developers, no page date shown.
  4. U.S. Securities and Exchange Commission, company concept response for Apple Inc., NetIncomeLoss, live data.
  5. Tan Solutions, US Company Filings and Financials, test run of 13 November 2026, verified 14 November 2026.

Next steps

Related guides