Form 10-KT Files Dataset

The Form 10-KT Files Dataset is a document-level corpus of every transition report filed on Form 10-KT or Form 10-KT/A by domestic Exchange Act registrants that change their fiscal year-end. A single record is one EDGAR submission — one complete transition report (or amendment), identified by a single accession number, packaged as a filing folder containing a structured metadata.json index plus every human-readable document that accompanied the original EDGAR submission (primary transition-report body and each numbered exhibit). The underlying form is filed pursuant to Section 13 or Section 15(d) of the Securities Exchange Act of 1934 and required by Rules 13a-10 and 15d-10 when the transition period exceeds six months. Coverage begins with EDGAR's earliest 10-KT submissions on 1995-03-01 and runs through present, with monthly ZIP containers keyed as YYYY/YYYY-MM.zip. The dataset is distributed in ZIP containers and carries TXT, JSON, HTML, and PDF file types, with inline XBRL facts preserved directly inside modern primary documents.

Update Frequency
Daily
Updated at
2026-05-01
Earliest Sample Date
1995-03-01
Total Size
97.9 MB
Total Records
4,517
Container Format
ZIP
Content Types
TXT, JSON, HTML, PDF
Form Types
10-KT, 10-KT/A

Dataset APIs

Programmatically retrieve the full list of dataset archive files, download URLs and dataset metadata.

Dataset Index JSON API

Download the entire dataset as a single archive file.

Download Entire Dataset:

Download a single container file (e.g. monthly archive) from the dataset.

Download Single Container:

Dataset Files

259 files · 97.9 MB
Download All
2026-04.zip53.7 KB3 records
2026-03.zip1.3 MB51 records
2026-02.zip473.7 KB20 records
2025-11.zip617.1 KB27 records
2025-08.zip542.0 KB24 records
2025-07.zip260.1 KB10 records
2025-06.zip41.8 KB3 records
2025-05.zip269.9 KB7 records
2025-04.zip172.0 KB15 records
2025-03.zip2.2 MB48 records
2024-12.zip45.4 KB3 records
2024-11.zip56.2 KB6 records
2024-10.zip383.8 KB6 records
2024-09.zip169.8 KB5 records
2024-08.zip476.8 KB14 records
2024-04.zip352.5 KB13 records
2024-03.zip1.5 MB40 records
2023-12.zip61.7 KB3 records
2023-09.zip59.8 KB5 records
2023-08.zip190.8 KB6 records
2023-05.zip506.0 KB17 records
2023-04.zip280.4 KB16 records
2023-03.zip1.3 MB37 records
2023-02.zip1.5 MB15 records
2023-01.zip121.8 KB4 records
2022-09.zip276.3 KB11 records
2022-08.zip274.8 KB7 records
2022-07.zip76.4 KB4 records
2022-06.zip202.5 KB5 records
2022-05.zip404.6 KB11 records
2022-04.zip248.2 KB10 records
2022-03.zip910.2 KB44 records
2022-02.zip1.7 MB54 records
2022-01.zip817.5 KB22 records
2021-12.zip171.4 KB5 records
2021-11.zip754.5 KB21 records
2021-10.zip81.2 KB4 records
2021-09.zip73.2 KB4 records
2021-06.zip798.1 KB36 records
2021-05.zip227.8 KB11 records
2021-04.zip133.0 KB9 records
2021-03.zip1.4 MB72 records
2021-02.zip430.8 KB6 records
2021-01.zip65.4 KB5 records
2020-12.zip61.2 KB6 records
2020-11.zip61.8 KB6 records
2020-10.zip115.4 KB21 records
2020-09.zip244.5 KB15 records
2020-08.zip169.2 KB16 records
2020-07.zip12.6 KB6 records
2020-06.zip54.0 KB3 records
2020-05.zip128.0 KB11 records
2020-04.zip289.7 KB16 records
2020-03.zip1.5 MB44 records
2019-12.zip225.6 KB24 records
2019-11.zip108.8 KB3 records
2019-07.zip121.5 KB8 records
2019-05.zip71.8 KB4 records
2019-04.zip83.8 KB6 records
2019-03.zip2.6 MB130 records
2019-02.zip919.8 KB28 records
2019-01.zip523.9 KB14 records
2018-12.zip254.7 KB5 records
2018-11.zip318.6 KB21 records
2018-10.zip179.4 KB14 records
2018-09.zip216.8 KB10 records
2018-08.zip74.1 KB6 records
2018-07.zip208.2 KB13 records
2018-06.zip510.1 KB27 records
2018-05.zip359.0 KB21 records
2018-04.zip672.7 KB28 records
2018-03.zip1.8 MB67 records
2018-02.zip125.0 KB9 records
2017-11.zip156.9 KB7 records
2017-10.zip66.3 KB6 records
2017-07.zip50.4 KB3 records
2017-06.zip143.3 KB6 records
2017-04.zip673.2 KB28 records
2017-03.zip1.5 MB50 records
2017-02.zip1.1 MB30 records
2016-12.zip56.2 KB3 records
2016-11.zip30.9 KB4 records
2016-10.zip353.5 KB12 records
2016-09.zip397.0 KB22 records
2016-08.zip97.6 KB4 records
2016-07.zip102.9 KB4 records
2016-06.zip340.2 KB23 records
2016-04.zip855.7 KB37 records
2016-03.zip2.6 MB147 records
2016-02.zip835.4 KB30 records
2016-01.zip310.7 KB7 records
2015-12.zip125.2 KB11 records
2015-11.zip387.3 KB18 records
2015-09.zip214.7 KB7 records
2015-07.zip58.8 KB5 records
2015-06.zip218.5 KB7 records
2015-05.zip166.6 KB16 records
2015-04.zip221.6 KB16 records
2015-03.zip1.7 MB65 records
2015-02.zip468.9 KB14 records
2014-12.zip245.1 KB10 records
2014-11.zip332.3 KB12 records
2014-10.zip211.7 KB15 records
2014-09.zip1.1 MB49 records
2014-08.zip285.7 KB10 records
2014-07.zip130.5 KB10 records
2014-06.zip5.6 KB1 records
2014-05.zip749.0 KB27 records
2014-04.zip926.8 KB41 records
2014-03.zip2.1 MB88 records
2014-02.zip1.5 MB60 records
2013-12.zip190.6 KB5 records
2013-07.zip347.1 KB17 records
2013-06.zip454.1 KB23 records
2013-05.zip274.1 KB15 records
2013-04.zip666.8 KB39 records
2013-03.zip1.7 MB58 records
2013-02.zip1.5 MB37 records
2013-01.zip346.1 KB17 records
2012-12.zip100.6 KB5 records
2012-11.zip252.8 KB10 records
2012-10.zip78.4 KB3 records
2012-09.zip262.0 KB13 records
2012-08.zip80.8 KB5 records
2012-07.zip39.8 KB6 records
2012-06.zip260.3 KB11 records
2012-05.zip596.1 KB23 records
2012-04.zip447.4 KB19 records
2012-03.zip1.2 MB49 records
2012-02.zip532.4 KB33 records
2011-12.zip219.8 KB13 records
2011-11.zip142.4 KB6 records
2011-09.zip185.3 KB11 records
2011-08.zip222.1 KB12 records
2011-07.zip121.7 KB6 records
2011-05.zip335.2 KB19 records
2011-04.zip385.2 KB28 records
2011-03.zip2.7 MB103 records
2011-02.zip605.4 KB31 records
2011-01.zip106.6 KB9 records
2010-12.zip298.0 KB13 records
2010-11.zip407.1 KB28 records
2010-10.zip65.0 KB3 records
2010-09.zip46.7 KB5 records
2010-08.zip31.4 KB6 records
2010-07.zip30.7 KB4 records
2010-06.zip80.9 KB5 records
2010-05.zip172.7 KB8 records
2010-04.zip1.0 MB47 records
2010-03.zip1.1 MB50 records
2010-01.zip195.3 KB11 records
2009-12.zip127.4 KB10 records
2009-11.zip484.0 KB31 records
2009-08.zip61.7 KB3 records
2009-07.zip101.5 KB5 records
2009-06.zip82.7 KB5 records
2009-05.zip81.7 KB5 records
2009-03.zip1.0 MB48 records
2008-10.zip149.2 KB17 records
2008-07.zip227.1 KB9 records
2008-06.zip184.9 KB22 records
2008-04.zip214.7 KB10 records
2008-03.zip332.0 KB14 records
2008-02.zip550.0 KB30 records
2008-01.zip103.0 KB6 records
2007-12.zip123.2 KB11 records
2007-11.zip507.8 KB33 records
2007-09.zip106.9 KB5 records
2007-08.zip206.9 KB13 records
2007-05.zip205.1 KB12 records
2007-04.zip205.6 KB11 records
2007-03.zip211.0 KB8 records
2007-02.zip525.1 KB26 records
2007-01.zip83.8 KB4 records
2006-12.zip112.5 KB8 records
2006-11.zip85.0 KB5 records
2006-10.zip300.9 KB8 records
2006-09.zip294.6 KB19 records
2006-08.zip16.9 KB5 records
2006-07.zip82.5 KB4 records
2006-06.zip625.4 KB19 records
2006-05.zip20.2 KB5 records
2006-04.zip25.6 KB5 records
2006-03.zip1.5 MB75 records
2005-12.zip53.0 KB5 records
2005-07.zip130.9 KB6 records
2005-06.zip97.2 KB6 records
2005-04.zip92.1 KB6 records
2005-03.zip594.6 KB35 records
2005-02.zip136.1 KB9 records
2005-01.zip255.6 KB18 records
2004-12.zip555.5 KB32 records
2004-04.zip36.3 KB3 records
2004-03.zip476.1 KB100 records
2004-02.zip158.7 KB11 records
2003-12.zip199.6 KB8 records
2003-11.zip329.9 KB19 records
2003-09.zip393.2 KB21 records
2003-08.zip364.3 KB17 records
2003-07.zip122.6 KB10 records
2003-06.zip285.8 KB10 records
2003-05.zip426.0 KB19 records
2003-04.zip815.1 KB26 records
2003-03.zip1.8 MB80 records
2003-01.zip276.8 KB19 records
2002-12.zip175.0 KB6 records
2002-11.zip256.3 KB16 records
2002-10.zip20.0 KB1 records
2002-09.zip246.6 KB9 records
2002-06.zip203.6 KB4 records
2002-05.zip108.9 KB3 records
2002-04.zip1.5 MB74 records
2002-03.zip324.3 KB12 records
2002-02.zip72.5 KB7 records
2002-01.zip62.9 KB1 records
2001-10.zip188.9 KB13 records
2001-07.zip122.5 KB7 records
2001-05.zip68.2 KB6 records
2001-04.zip418.4 KB19 records
2001-03.zip529.6 KB26 records
2000-12.zip125.3 KB7 records
2000-11.zip16.4 KB2 records
2000-10.zip119.4 KB4 records
2000-09.zip188.2 KB11 records
2000-08.zip73.0 KB2 records
2000-06.zip88.0 KB7 records
2000-04.zip62.0 KB7 records
2000-03.zip236.3 KB14 records
1999-12.zip43.7 KB2 records
1999-10.zip124.1 KB12 records
1999-05.zip174.3 KB10 records
1999-04.zip93.2 KB4 records
1999-03.zip488.0 KB34 records
1998-11.zip83.8 KB8 records
1998-10.zip38.5 KB5 records
1998-09.zip128.0 KB2 records
1998-07.zip282.5 KB21 records
1998-06.zip141.7 KB8 records
1998-05.zip51.6 KB2 records
1998-04.zip207.6 KB13 records
1998-03.zip466.0 KB31 records
1998-01.zip56.8 KB11 records
1997-12.zip79.8 KB7 records
1997-10.zip52.7 KB3 records
1997-09.zip52.5 KB3 records
1997-07.zip163.1 KB12 records
1997-06.zip52.1 KB5 records
1997-05.zip35.0 KB6 records
1997-04.zip130.9 KB7 records
1997-03.zip54.8 KB5 records
1997-01.zip58.7 KB6 records
1996-10.zip10.7 KB2 records
1996-04.zip117.6 KB12 records
1996-03.zip98.1 KB10 records
1996-02.zip188.7 KB7 records
1995-12.zip80.2 KB14 records
1995-08.zip30.9 KB3 records
1995-06.zip94.8 KB9 records
1995-03.zip39.8 KB2 records

What This Dataset Contains

The dataset captures every domestic transition report on the annual-form track — Form 10-KT and its amendment variant Form 10-KT/A — submitted to EDGAR from March 1995 to present. Form 10-KT is the annual-report analogue used when a registrant changes its fiscal year-end. Filed pursuant to Section 13 or 15(d) of the Securities Exchange Act of 1934 and required by Rules 13a-10 and 15d-10, it covers the "stub" transition period between the close of the prior fiscal year and the opening of the newly adopted fiscal year. Because 10-KT inherits the disclosure regime of Form 10-K rather than defining its own, the content of a 10-KT mirrors the four-part 10-K architecture — business description, risk factors, properties, legal proceedings, MD&A, audited financial statements, controls-and-procedures certifications, governance and compensation matters, and exhibits — with two substantive differences: the reporting period is the transition window rather than a full fiscal year, and comparative prior-period columns in the financial statements are re-cast to align with that stub. Form 10-KT/A is the amendment variant, used to restate, correct, or supplement an earlier transition report; it typically repeats only the amended portions plus reissued certifications, and its cover page and title prose explicitly flag the amended status.

Each record captures the filing at two layers simultaneously: a machine-readable normalised metadata layer describing every component of the submission, and a document layer preserving the issuer's as-filed narrative, financial, and certification content. The dataset is delivered as ZIP containers organised by filing month (YYYY/YYYY-MM.zip), grouped by the filing's filedAt month, and the file types found across records are TXT, JSON, HTML, and PDF. The HTML set covers both modern XHTML / inline-XBRL filings and legacy <HTML>-wrapped documents that EDGAR accepted from the late 1990s onward; TXT covers pre-HTML plain-text filings and occasional exhibits submitted as text; JSON is exclusively the metadata.json index; and PDF covers exhibit attachments filed in PDF form. Binary image attachments and XBRL linkbase/schema files are intentionally omitted from the packaged record but are catalogued with resolvable URLs in metadata.json.

Content Structure of a Single Record

A single record in the Form 10-KT Files Dataset is one EDGAR submission filed on Form 10-KT or Form 10-KT/A, identified by a single accession number. On disk the record is materialised as a filing folder whose name is the 18-digit accession number with the dashes stripped, nested inside a monthly ZIP keyed as YYYY/YYYY-MM.zip. Each record is a flat directory of sibling files with no nested subfolders. The files fall into three roles:

  1. metadata.json — a single JSON document enumerating the filing's header facts, filer entities, every attachment EDGAR recognised (human-readable and technical), and canonical links back to EDGAR.
  2. The primary transition-report document — one HTML file (or plain text in older filings) carrying the body of the 10-KT or 10-KT/A. In modern filings this is served as inline XBRL (iXBRL): valid XHTML with embedded <ix:header>, <ix:hidden>, <ix:references>, <ix:resources>, <ix:nonNumeric>, and <ix:nonFraction> fact tags that bind DEI and financial-statement values to the narrative text.
  3. Exhibit documents — one file per exhibit attached to the filing, most commonly Sarbanes-Oxley certifications (EX-31.1, EX-31.2, EX-32.1, EX-32.2), and, depending on the issuer, material contracts, subsidiary lists (EX-21), consents of independent auditors (EX-23), computation-of-ratios tables, corporate-governance documents, clawback policies (EX-97), and other Item 601 exhibits carried over from the 10-K exhibit index.

Every document file preserves the EDGAR SGML envelope around its payload. Opening a file reveals a <DOCUMENT> wrapper with <TYPE>, <SEQUENCE>, <FILENAME>, and <DESCRIPTION> tags followed by a <TEXT> block containing the actual HTML (or plain-text) body between <HTML>…</HTML> tags. These SGML tags mirror the corresponding entry in metadata.json's documentFormatFiles array one-to-one, allowing each on-disk file to be linked back to its role in the submission.

metadata.json — field-level breakdown

metadata.json is the authoritative index for the record. Its top-level keys are:

  • formType — either 10-KT or 10-KT/A.
  • accessionNo — the canonical dashed accession number (e.g. 0001376474-25-000520).
  • description — the SEC-supplied submission description, typically "Form 10-KT - Transition reports [Rule 13a-10 or 15d-10]", with [Amend] appended on amendments.
  • filedAt — ISO-8601 filing timestamp with the EDGAR timezone offset (e.g. 2025-06-09T15:51:14-04:00).
  • periodOfReport — the transition period end date in YYYY-MM-DD form.
  • linkToFilingDetails — URL to the primary document on EDGAR.
  • linkToTxt — URL to the "Complete submission text file", which is the concatenated SGML submission containing every document.
  • linkToHtml — URL to the EDGAR filing-index page (-index.htm).
  • linkToXbrl — URL to a separately-attached XBRL instance when one exists; empty when the filing uses inline XBRL and the facts live inside the primary HTML.
  • id — an opaque 32-hex-character dataset-internal identifier.
  • documentFormatFiles — an ordered array of every attachment in the submission. Each entry carries sequence, size (raw byte count as a string), documentUrl, description, and type. The first sequence is the main filing body (type = 10-KT or 10-KT/A); subsequent entries are exhibits (EX-31.1, EX-32.1, EX-21, EX-23, etc.), graphic attachments (GRAPHIC), and a terminal pseudo-entry with blank sequence/type that points at the complete-submission text file.
  • dataFiles — the XBRL technical artefacts, typed EX-101.INS / XML (extracted instance document), EX-101.CAL, EX-101.DEF, EX-101.LAB, EX-101.PRE, and EX-101.SCH for the calculation, definition, label, and presentation linkbases and the extension schema.
  • entities — an array of registrant objects, one per filer on the submission. Per-entity keys include companyName, cik, type (the form type as listed in the header), fiscalYearEnd (four-digit MMDD), stateOfIncorporation, irsNo, fileNo, filmNo, act (commonly "34" for the Exchange Act), sic (SIC code and industry name, with HTML entity encoding preserved from the EDGAR header, so &amp; appears literally), and tickers (an array of associated ticker symbols that may include both current and historical symbols for the issuer).
  • seriesAndClassesContractsInformation — populated only for investment-company filers that use the EDGAR series/class framework; empty for operating-company 10-KT filers.

The array fields are particularly useful for extraction: documentFormatFiles lets a consumer map every .htm, .txt, or .pdf on disk to its declared exhibit type, and dataFiles provides a complete reference to the XBRL linkbases even though those files are not bundled inside the ZIP.

Section-by-section content of the 10-KT body

The primary document follows the four-part architecture of Form 10-K, adapted to the transition period:

In a 10-KT/A amendment, the body is usually trimmed to an Explanatory Note followed by only the amended Items, a re-signed signature block, and a reissued Item 15 exhibit index; certifications under Exhibits 31 and 32 are always reissued even when the substantive amendment is narrow.

Exhibit layer

Each exhibit is a separate file in the filing folder, written as a standalone <DOCUMENT> block with a TYPE such as EX-31.1, EX-32.1, EX-21, EX-23, EX-10.x, EX-3.i, EX-4, EX-97 (clawback policy, required since 2023), or EX-101.* for the XBRL linkbases (catalogued in metadata but not bundled). The two exhibits that appear in virtually every post-2002 10-KT record are:

  • Exhibit 31Sarbanes-Oxley Section 302 certification(s), one per principal officer, with the five-paragraph attestation mandated by Rule 13a-14(a)/15d-14(a): the officer has reviewed the report; based on their knowledge it does not contain untrue statements or omissions; the financial statements fairly present; they are responsible for establishing and maintaining disclosure controls and ICFR; and they have disclosed to the auditors and audit committee any significant deficiencies, material weaknesses, or fraud. In 10-KT/A amendments the attestation language is adjusted to refer to the "amended Transition Report on Form 10-KT/A".
  • Exhibit 32Sarbanes-Oxley Section 906 certification(s), furnished rather than filed, in which each officer certifies under 18 U.S.C. § 1350 that the report fully complies with Section 13(a) or 15(d) and that the information fairly presents the financial condition and results of operations.

Other common exhibits include EX-21 (list of subsidiaries), EX-23 (consent of independent registered public accounting firm), EX-24 (powers of attorney), EX-4 (description of securities registered under Section 12), EX-10 (material contracts), EX-3 (charter and bylaws), and EX-97 (compensation-recovery / clawback policy).

What is included in the dataset

The dataset copy of a record includes:

  • metadata.json with every field described above.
  • The primary 10-KT or 10-KT/A document (HTML, or plain text in older submissions).
  • All exhibit documents in their human-readable form (HTML, TXT, or PDF, depending on how the issuer filed them). PDFs are retained when they were the as-filed format for an exhibit — for example, graphical or scanned exhibits submitted as PDF rather than HTML.
  • The original EDGAR SGML <DOCUMENT> envelope around every document file, so the TYPE, SEQUENCE, FILENAME, and DESCRIPTION tags survive on disk.

What is excluded or left as external references

Two categories of attachment are catalogued in metadata.json but not physically bundled:

  • Image and graphic files — any GRAPHIC-typed entry in documentFormatFiles (JPEG, GIF, PNG, etc. used for logos, charts, signatures, maps, or exhibit graphics) is omitted from the packaged record. The entry still appears in documentFormatFiles with its full documentUrl, so the image can be retrieved live from EDGAR when needed, but the ZIP itself carries only the text and HTML that reference those images.
  • XBRL technical artefacts — the calculation, definition, label, and presentation linkbases (EX-101.CAL, EX-101.DEF, EX-101.LAB, EX-101.PRE), the extension schema (EX-101.SCH), and the extracted XBRL instance document (*_htm.xml) are enumerated in metadata.json's dataFiles array with full URLs but not bundled. For filings that use inline XBRL, the structured facts nevertheless remain present inside the primary document, because iXBRL tags live directly in the XHTML body.

Also structurally separate from the record is the proxy statement that Part III items are frequently incorporated by reference to — that content is filed as a DEF 14A submission under a different accession number and is not included here.

Historical evolution of required content

The 10-KT inherits every structural change the SEC has made to Form 10-K, so the required Items and their contents have shifted materially over the life of the form:

  • 1995–2002 — Part III disclosures were lighter, Item 9A "Controls and Procedures" did not yet exist, and officer certifications were not required. Early transition reports also relied heavily on EDGAR's pre-HTML text conventions.
  • 2002 (Sarbanes-Oxley) — Exhibit 31 (Section 302) and Exhibit 32 (Section 906) certifications became mandatory attachments, and Item 9A began requiring management's evaluation of disclosure controls.
  • 2004–2010 (Section 404 rollout) — management's report on internal control over financial reporting, and, for non-smaller-reporting-company filers, the auditor's attestation on ICFR, were phased into Item 9A.
  • 2005 — Risk Factors (Item 1A) and Unresolved Staff Comments (Item 1B) were added to Part I under the Securities Offering Reform.
  • 2010 — Mine Safety Disclosures (Item 4 of Part I) were added following the Dodd-Frank Act.
  • 2017PCAOB Release No. 2017-001 required the auditor's report to disclose auditor tenure and, for large accelerated filers starting in 2019, Critical Audit Matters.
  • 2020 (FAST Act Modernization) — Item 1 Business and Item 1A Risk Factors were streamlined with a principles-based approach and a risk-factor summary requirement when risk factors exceed 15 pages.
  • 2021 — Item 6 Selected Financial Data was rescinded and marked [Reserved]; Item 9C Disclosure Regarding Foreign Jurisdictions that Prevent Inspections was added under the Holding Foreign Companies Accountable Act.
  • 2023 — Item 1C Cybersecurity was added to Part I (risk management, strategy, governance, and disclosure of material cybersecurity incidents), and Exhibit 97 (clawback / compensation-recovery policy) became a required exhibit under Dodd-Frank-driven listing standards.

A 10-KT filed in any given year conforms to the Item set in effect at that time; amendments (10-KT/A) generally conform to the rules in effect when the amendment is filed, though they may restate portions of an earlier-era report.

Historical evolution of data format

The presentation format of a 10-KT record has gone through several regimes:

  • 1995 through the late 1990s — filings were submitted to EDGAR as plain ASCII text, with financial tables rendered in monospaced column alignment. The SGML <DOCUMENT> wrapper already existed; the <TEXT> payload was ASCII rather than HTML. Records from this era appear in the dataset as .txt files.
  • Late 1990s through the 2000s — HTML became the predominant submission format, wrapped inside the same <DOCUMENT> envelope with <TEXT><HTML>…</HTML></TEXT>. Tables, headings, and exhibit numbering became visually formatted rather than column-aligned text.
  • 2009–2017 (phased XBRL mandate) — the SEC's interactive-data rules required XBRL tagging of financial statements and, later, footnotes, attached as separate EX-101.* files alongside the human-readable HTML. In this era the HTML body and the XBRL instance were distinct files; the dataset bundles the HTML and points to the XBRL via dataFiles URLs.
  • 2019 onward (inline XBRL) — the SEC's inline XBRL rule required financial-statement facts and cover-page data to be tagged directly inside the primary HTML using the iXBRL namespace. In modern records the primary document is an XHTML file that carries <ix:header>, <ix:hidden>, <ix:references>, and <ix:resources> blocks, plus <ix:nonNumeric> and <ix:nonFraction> tags around the text and numeric values, binding DEI facts (entity CIK, document fiscal year focus, fiscal period focus, amendment flag, period end date) and financial facts to <xbrli:context> elements that define the transition-period and comparative reporting ranges. For these filings linkToXbrl is an empty string because no separate XBRL instance is attached; the extracted _htm.xml companion is still listed in dataFiles but not bundled.

Filename conventions have likewise evolved. Modern primary documents follow the pattern {ticker-or-slug}-{YYYYMMDD}.htm, where the date is the transition-period end. Exhibits use issuer-chosen slugs such as ex31z1.htm, ex32_1.htm, or {ticker}_ex21.htm. Older filings use less structured names and frequently bundle multiple Items into a single .txt file.

Interpretation and extraction notes

Several nuances matter when reading or extracting from these records:

  • Amendments versus originals — a 10-KT/A record may contain only the amended portions of the original transition report. To reconstruct the full as-amended report, a consumer must also retrieve the original 10-KT accession (identified by matching cik and a nearby periodOfReport).
  • Incorporation by reference — Part III Items are routinely incorporated by reference to the forthcoming definitive proxy statement, so the 10-KT body itself may contain only a pointer for Items 10 through 14. This is a feature of the source filing, not of the dataset packaging.
  • SGML wrappers are preserved — every document file on disk begins with <DOCUMENT><TYPE>…<SEQUENCE>…<FILENAME>…<DESCRIPTION>…<TEXT> before the HTML body. Parsers must either strip this envelope or use the TYPE tag to classify the file. The tag values correspond one-to-one with documentFormatFiles[].type entries in metadata.json, which is the cleanest programmatic way to enumerate exhibits.
  • Inline XBRL extraction — for iXBRL filings every tagged fact can be pulled from the primary HTML without consulting the external linkbases, though the linkbases (referenced via dataFiles URLs) are still needed to resolve calculation relationships, preferred labels, and presentation hierarchy.
  • Image references in the body — the primary HTML may contain <img> tags pointing at filenames listed in documentFormatFiles as GRAPHIC entries that are not packaged in the ZIP. These references will appear as broken links if the HTML is opened offline; the documentUrl in the corresponding metadata entry resolves to the live EDGAR image.
  • Multiple filers on one submission — the entities array can carry more than one registrant when a parent and a subsidiary or multiple co-registrants file jointly; each entry carries its own CIK, SIC, fiscal-year-end, and ticker list.
  • Ticker history — the tickers array within an entity may enumerate both the currently traded symbol and historical symbols, reflecting ticker changes across the issuer's reporting history.
  • Transition-period datingperiodOfReport is the end of the transition window, not a traditional fiscal year end; combined with the cover-page disclosure of the transition-period start date, it is the definitive way to date a 10-KT's reporting window. A fiscal-year change can be inferred by comparing entities[0].fiscalYearEnd against the periodOfReport month and day.
  • SIC encoding quirkentities[0].sic retains HTML entity encoding from the raw EDGAR header (e.g. 1311 Crude Petroleum &amp; Natural Gas); consumers should decode the string before display.
  • File-size semanticsdocumentFormatFiles[].size is a string of raw uncompressed byte counts and matches the uncompressed sizes observed in the ZIP central directory, making it safe to use for integrity checks without decompressing.
  • Complete-submission text file — the final blank-typed entry in documentFormatFiles and the linkToTxt URL both refer to EDGAR's concatenated SGML submission, which contains every document including the otherwise-excluded images and XBRL artefacts, and can be retrieved live when the full as-filed package is needed.

Who Files or Publishes This Dataset, and When

Who files the record

Each record is a single EDGAR submission of Form 10-KT (a transition report in lieu of an annual report) or Form 10-KT/A (an amendment). The filer is the domestic Exchange Act registrant whose fiscal year-end has changed, submitting under its own CIK. No affiliated person, subsidiary, officer, or director files a 10-KT on the registrant's behalf.

Eligible filers are the same population that otherwise files Form 10-K: domestic operating companies, holding companies, and other commercial issuers subject to Section 13(a) (by virtue of a class of securities registered under Section 12) or Section 15(d) (by virtue of an effective Securities Act registration statement). All domestic filer tiers appear:

Successor registrants (for example, a newly formed holding company inserted above an operating registrant) also appear when the successor adopts a different fiscal year than the predecessor.

Outside this population: Foreign private issuers report fiscal-year transitions on Form 20-F (or Form 40-F under MJDS), not 10-KT. Registered investment companies use the N-CSR / N-series framework. Asset-backed issuers on Form 10-K under Regulation AB are technically eligible but rarely file 10-KTs because ABS structures seldom change fiscal years.

What triggers the record

The record is event-driven, not periodic. A 10-KT is triggered when a Section 13(a) or 15(d) reporting company changes its fiscal year-end, under Rule 13a-10 and Rule 15d-10 of the Exchange Act. These rules govern the transition period — the span between the close of the prior fiscal year on the old calendar and the start of the new fiscal year on the new calendar.

The triggering act is the registrant's own decision (typically a board resolution) to adopt a new fiscal year-end. Common motivations include alignment with a new parent, post-acquisition conformity, industry-cycle alignment, or audit and tax scheduling. The obligation does not recur unless the registrant again changes its fiscal year.

Rules 13a-10 and 15d-10 determine which form covers the transition period based on its length:

  • One month or less: no separate transition report; the transition period is covered in the next annual report on Form 10-K.
  • More than one month but not more than six months: the registrant may file the transition report on Form 10-Q (as a transition report on that form). The transition-period audited financials are then picked up in the next Form 10-K filed on the new calendar.
  • More than six months: the registrant must file the transition report on Form 10-K — that is, as a Form 10-KT covering the transition period.

In practice, the records in this dataset therefore correspond to transition periods longer than six months, and the financial statements they contain are audited for that span. The first full fiscal quarter on the new calendar is a regular Form 10-Q, so a 10-KT typically coexists with ordinary 10-Qs rather than replacing them.

Regulatory framework

  • Section 13(a) / Section 15(d) of the Exchange Act — underlying annual reporting obligation.
  • Rule 13a-10 (Section 12 registrants) and Rule 15d-10 (Section 15(d) registrants) — transition-report mechanics, form choice, and deadline.
  • Form 10-K / 10-KT instructions — disclosure content: audited financial statements, MD&A, business, risk factors, properties, legal proceedings, market information, directors and officers, executive compensation, security ownership, related-party transactions, and principal accountant fees, adapted to the transition period.
  • Regulation S-X — form and content of the audited transition-period financial statements, including comparative periods.
  • Regulation S-K — non-financial disclosure items incorporated into Form 10-KT.
  • Sarbanes-Oxley Sections 302 and 906 — CEO/CFO certifications filed as exhibits, referenced to the transition period.
  • SOX Section 404 — ICFR reporting where applicable based on filer status, assessed as of the end of the transition period.

Part III disclosures may be incorporated by reference from the definitive proxy statement or information statement, in the same manner as for Form 10-K, if that statement is filed within 120 days of the end of the transition period.

Deadlines

Rules 13a-10(g) and 15d-10(g) set the 10-KT deadline by reference to the Form 10-K tiered schedule, measured from the last day of the transition period:

  • Large accelerated filers: 60 days after the end of the transition period.
  • Accelerated filers: 75 days after the end of the transition period.
  • Non-accelerated filers (including most smaller reporting companies): 90 days after the end of the transition period.

For transition periods of six months or less reported on Form 10-Q rather than 10-KT, Rules 13a-10 and 15d-10 generally set the deadline as the later of the otherwise-applicable 10-K deadline for the prior fiscal year or the 10-Q deadline for the transition period — but those filings are not part of this dataset.

Form 10-KT/A has no standalone statutory deadline; amendments are filed when the registrant determines that prior transition-period disclosure must be corrected, restated, supplemented, or brought into compliance. Common drivers: restatement of transition-period financials, late Part III when the proxy was not filed within 120 days, correction of exhibits or certifications, and responses to SEC staff comments.

Important distinctions

  • Successor issuers. After a reorganization or holding-company insertion, the successor files the 10-KT under its own CIK if it adopts a different fiscal year than the predecessor — even though the disclosure largely describes predecessor operations.
  • Business-combination-driven changes. Acquisitions (including reverse mergers) that conform one party's fiscal year to the other's are a frequent source of 10-KTs. The filer is the continuing Exchange Act registrant, not the acquired target.
  • 10-KT vs 10-K. A 10-KT covers only the transition period. The first full fiscal year on the new calendar is reported on a regular Form 10-K and is not included in this dataset.
  • 10-KT vs transition 10-Q. Transitions of six months or less are reported on Form 10-Q and do not appear here; transitions of one month or less require no separate transition report at all.
  • Back-to-back year-end changes. An issuer that changes fiscal year more than once files a separate 10-KT each time the six-month trigger is met.
  • 10-KT/A. Amendments are filed by the same registrant under the same CIK and are separate EDGAR accessions; each appears as its own record rather than being merged with the original.
  • Filer vs persons described. The registrant is the sole filer. Named executives, directors, and beneficial owners are subjects of Part III disclosure; their personal reporting obligations (Forms 3, 4, 5, Schedule 13D/G) remain separate.
  • Non-domestic and non-operating issuers. FPIs (Form 20-F / 40-F) and registered investment companies (Form N-CSR) fall outside this reporting path entirely.
  • Earliest records. EDGAR electronic 10-KT records begin in March 1995, reflecting the EDGAR phase-in. The underlying Rule 13a-10 / 15d-10 transition-report obligation predates EDGAR; earlier transition reports exist only in the SEC's paper archive and are not in this dataset.

How This Dataset Differs From Similar Datasets or Filings

Form 10-KT is an event-driven periodic report filed only when a domestic registrant changes its fiscal year-end. That narrow trigger places it adjacent to several standard filing datasets without being interchangeable with any of them.

Form 10-K annual report files

Form 10-K is the closest structural relative. A 10-KT reuses the same Regulation S-K narrative architecture (business, risk factors, MD&A, legal proceedings, controls) and the same Regulation S-X audited-financial-statement requirements as a 10-K. The decisive difference is the reporting window: a 10-K covers a full 12-month fiscal year; a 10-KT covers the "stub" transition period between the old and new fiscal year-ends. Under Rule 13a-10 / 15d-10, transition financial statements cover a non-standard 7-to-12-month span, comparative-period presentation follows the transition-report rules rather than ordinary annual comparison, and the filing deadline is anchored to the end of the transition period. Scale differs by orders of magnitude: the 10-KT corpus since March 1995 is a small fraction of the 10-K dataset corpus over the same horizon.

Form 10-KT/A amendments (included in this dataset)

10-KT/A filings sit in the same dataset as original 10-KTs. The "/A" amends a prior transition report to restate financials, correct disclosures, add omitted exhibits (auditor consents, Part III proxy carryover), or respond to staff comments. It does not extend the transition window. Treat originals and each /A as separate records for point-in-time analysis; follow the accession chain to the latest /A for the authoritative version.

Form 10-QT (short-period transition report)

Form 10-QT is the quarterly-style transition analog. The choice between 10-KT and 10-QT is governed by transition length: periods longer than six months go on Form 10-KT with audited, annual-style disclosure; periods of six months or less go on Form 10-QT with unaudited interim-style financials. The 10-QT dataset therefore excludes every short-period fiscal-year shift. Researchers tracking a specific registrant's year-end change should check both populations.

Form 20-F (foreign private issuer transition reports)

Foreign private issuers changing fiscal year-end file a transition report on Form 20-F itself, using the transition-report checkbox on the cover page, not on Form 10-KT. The 10-KT dataset is consequently domestic-only. Cross-border fiscal-year-change research requires joining this dataset with a 20-F dataset filtered on transition indicators.

Form 8-K Item 5.03 (upstream announcement)

Item 5.03 of Form 8-K ("Amendments to Articles of Incorporation or Bylaws; Change in Fiscal Year") is the real-time disclosure that announces a year-end change, typically shortly after board approval. The 8-K announces; the 10-KT delivers the resulting audited financials and full transition-period disclosure months later. The two are sequential rather than overlapping, and an 8-K Item 5.03 to 10-KT join is the natural way to trace the full lifecycle of a fiscal-year change (including cases that resolve into a 10-QT or 20-F instead).

Investment company reports (Form N-CSR / N-CSRS)

Registered investment companies report under the Investment Company Act regime, not Section 13/15(d), and file certified shareholder reports on N-CSR / Form N-CSRS. A fund changing its fiscal year-end adjusts its N-CSR cycle; it does not file a 10-KT. This is a common pitfall when assembling an "all annual reports" corpus: the 10-KT dataset is operating-company only.

XBRL and structured financial-fact datasets

XBRL datasets deliver parsed, tagged numeric facts keyed to standardized concepts. The 10-KT files dataset is document-centric: original HTML, TXT, PDF, and JSON metadata as submitted to EDGAR. Because transition filings cover non-standard periods, some XBRL consumers mis-handle the period context, so the raw documents are often the more reliable source for period-sensitive narrative, exhibits, certifications, and auditor reports. The two dataset types complement rather than substitute for each other.

Boundary summary

Three constraints define this dataset and prevent substitution by any neighbor:

  1. Domestic Section 13(a) or 15(d) registrants only (excludes FPIs on Form 20-F and funds on N-CSR).
  2. Fiscal-year-end change with a transition period longer than six months (excludes Form 10-QT).
  3. Complete filing documents, not extracted facts or announcement events (distinct from XBRL datasets and 8-K Item 5.03 coverage).

No single alternative covers the same slice: Form 10-K datasets miss the transition window, 10-QT datasets cover shorter windows, 20-F datasets cover foreign issuers, 8-K 5.03 captures the announcement rather than the report, and XBRL strips out narrative and exhibits. The 10-KT files dataset is the authoritative document corpus for every domestic transition report on the annual-form track since EDGAR began.

Who Uses This Dataset

Form 10-KT documents the stub period created when a registrant changes its fiscal year-end. Because these transition reports break the normal annual cadence, the dataset supports technical workflows where the reporting discontinuity must be handled correctly.

Fundamental and equity research analysts

Analysts covering registrants that shift fiscal year-end rebuild comparable revenue, margin, and cash-flow series across the pre- and post-change regimes. Without the stub-period figures, annual models double-count or drop months of activity. They pull the audited income statement, balance sheet, and cash-flow statement from the primary HTML, plus the MD&A discussion of seasonality effects, to restate history onto the new calendar.

M&A and due-diligence teams

Buy-side diligence, corporate development, and sell-side advisers use 10-KTs when a target previously changed its fiscal year-end, often tied to an earlier acquisition or spin-off. They rely on the transition-period financials, the business-description and risk-factor sections explaining the calendar change, and any merger or reorganization exhibits to reconstruct a clean pre-transaction history for valuation and synergy work.

Auditors and technical accounting specialists

Audit and technical accounting teams use prior 10-KTs as precedent when advising clients contemplating their own fiscal-year change or reviewing a newly acquired entity's transition accounting. They focus on the audit report, basis-of-presentation footnote, comparative-period disclosures, restatement language, and the Section 302 and 906 certifications in the exhibit index. The corpus feeds internal knowledge bases covering transition-period presentation, segment rollovers, goodwill testing dates, and tax provision mechanics.

Quantitative researchers and financial data engineers

Quants building factor libraries, event studies, and backtests need fiscal-year changes flagged explicitly to prevent corruption of TTM growth, YoY comparisons, and accruals metrics. The metadata.json fields (accession, period-of-report, CIK, form type) allow programmatic detection of transition events, while the primary HTML provides the stub-period figures needed to bridge adjacent 10-Ks.

Forensic accountants and restatement researchers

Forensic accountants watch 10-KT/A amendments closely, since an amended transition report often signals unusual accounting issues, late auditor changes, or corrections that carry into the next full year. They diff the original 10-KT against the 10-KT/A, inspect amended auditor reports and re-issued certifications in the exhibits, and review the explanatory notes describing each change.

Academic accounting and finance researchers

Faculty, doctoral students, and research staff at policy institutions study motivations behind fiscal-year changes, including earnings management around the transition window, calendar alignment with an acquirer, and tax or seasonality rationales. The panel from March 1995 to present supports empirical designs linking fiscal-year changes to earnings quality, analyst coverage, or market reactions. They parse MD&A and financial statements programmatically and use metadata.json for sample construction.

Securities lawyers and disclosure counsel

External counsel and in-house legal teams drafting a client's 10-KT use the corpus as a drafting library. They review how peers framed risk factors, business descriptions, legal proceedings, and forward-looking statements in a transition context, and how Section 302/906 certifications, auditor consents, and Item 9A internal-control disclosures were tailored for a stub period. The primary HTML and exhibit index carry most of the weight here.

SEC reporting and compliance teams

Internal reporting teams at issuers contemplating a fiscal-year change benchmark reporting structure, exhibit lists, certification language, and cover-page conventions against peer filers, with particular attention to Rule 13a-10 and 15d-10 compliance. The dataset supports producing draft filings that align with established practice before outside counsel review.

Credit analysts and covenant-monitoring teams

Credit analysts at banks, private-debt funds, and rating groups use 10-KTs to reconcile covenants defined on a fiscal-year basis when a borrower shifts its year-end. Leverage ratios, fixed-charge coverage, and minimum-EBITDA tests depend on trailing-period definitions that must be rebuilt around the transition. They map the audited statements and MD&A from the primary document to covenant definitions in the underlying credit agreement, usually alongside the next full-year 10-K.

Corporate governance and proxy analysts

Governance analysts examine 10-KTs when a fiscal-year change coincides with board turnover, change in control, or reorganization. Relevant content includes Part III incorporation language, executive compensation disclosures for the stub period, and exhibits covering material agreements and governance documents.

LLM and RAG developers for financial text

Teams building retrieval-augmented systems and finance-domain language models use the dataset to ensure transition-period reporting is represented correctly. Because 10-KTs are rare relative to 10-Ks, naive scrapers miss them, producing hallucinated annual figures or gaps in company timelines. Clean form-type labeling in metadata.json and consistent text, HTML, and PDF representations simplify chunking, embedding, and evaluation.

Specific Use Cases

Form 10-KT records support a small set of concrete workflows built around the fiscal-year-end change, the audited stub-period financials, and the reissued certifications and exhibits. The examples below tie each workflow to specific fields in metadata.json, specific Items in the primary HTML body, and specific exhibits.

Bridging annual time series across a fiscal-year change

Equity and credit analysts rebuild revenue, EBITDA, and cash-flow series for issuers whose year-end shifted. The workflow pulls entities[0].fiscalYearEnd and periodOfReport from metadata.json to detect the discontinuity, then extracts Item 7 MD&A and the Item 8 audited statements of operations and cash flows from the primary HTML to splice the stub months onto the adjacent 10-Ks. Output is a continuous monthly or quarterly series with the transition window annotated, usable directly in DCF models and covenant roll-forwards.

Programmatic detection and flagging of fiscal-year changes in factor libraries

Quant teams filter EDGAR submissions by formType equal to 10-KT or 10-KT/A and index the resulting cik, accessionNo, and periodOfReport values into a point-in-time events table. This flag is then joined against TTM growth, accrual, and YoY factor calculations so that transition filings are either skipped or rescaled, preventing mechanical double-counting when a registrant reports two "annual" filings within a single calendar year.

Diff-based review of 10-KT/A amendments for restatement research

Forensic accountants pair each 10-KT/A accession with the prior original 10-KT for the same cik and adjacent periodOfReport, then diff the Item 8 financial statements, the Item 9A controls disclosure, the Item 9 changes-in-accountants language, and the reissued EX-31 and EX-32 certifications. The Explanatory Note at the top of the amended body and any added EX-23 auditor consent are tagged to build a structured restatement log covering scope, dollar impact, and auditor involvement.

Precedent library for drafting a transition report

Disclosure counsel and in-house SEC reporting teams preparing a client's own fiscal-year change query the corpus by SIC code (entities[0].sic) and transition length (derived from periodOfReport minus the cover-page transition-period start date) to retrieve peer templates. They extract the cover-page checkbox block, the Item 1A risk-factor framing of the calendar change, the Item 7 MD&A basis-of-presentation paragraph, and the EX-31/EX-32 wording that substitutes "Transition Report" for "Annual Report" — producing a reusable drafting kit aligned with Rule 13a-10 / 15d-10 practice.

Auditor-tenure and PCAOB ID extraction across transition reports

Audit-quality researchers and procurement teams benchmarking auditor relationships parse the independent auditor's report inside Item 8 to capture PCAOB firm ID, auditor tenure disclosure, and any Critical Audit Matters, and cross-reference the EX-23 consent for firm name changes. Joining that output with entities[0].cik and filedAt yields a panel of auditor-client pairs specifically at the moment of a fiscal-year change, which is a common trigger for auditor turnover.

Training and evaluation data for finance-domain LLMs

RAG and fine-tuning teams use the dataset as an explicit minority class in "annual report" corpora to cover the non-standard reporting window. Records are chunked using the SGML <DOCUMENT> envelope and the documentFormatFiles[].type field to split the primary 10-KT body from each EX-10 material contract, EX-21 subsidiary list, and EX-97 clawback policy. Inline XBRL <ix:nonFraction> tags in the primary HTML provide labeled numeric supervision for extraction evaluation on transition-period facts, where generic 10-K-trained models typically fail.

Dataset Access

Dataset Index JSON API: https://api.sec-api.io/datasets/form-10kt-files.json

This endpoint returns the dataset's metadata, the download URL for the full archive, and the complete list of monthly container files with per-container size, record count, updated timestamp, and download URL. Poll it to detect which containers changed in the most recent refresh and download only those containers on a daily basis. This endpoint does not require an API key.

Example response:

Example
1 {
2 "datasetId": "1f13365b-9ae0-69b1-8b6d-5aae9bbc3414",
3 "datasetDownloadUrl": "https://api.sec-api.io/datasets/form-10kt-files.zip",
4 "name": "Form 10-KT Files Dataset",
5 "updatedAt": "2026-04-15T12:11:47.148Z",
6 "earliestSampleDate": "1995-03-01",
7 "totalRecords": 4514,
8 "totalSize": 97855085,
9 "formTypes": ["10-KT", "10-KT/A"],
10 "containerFormat": "ZIP",
11 "fileTypes": ["TXT", "JSON", "HTML", "PDF"],
12 "containers": [
13 {
14 "downloadUrl": "https://api.sec-api.io/datasets/form-10kt-files/2026/2026-03.zip",
15 "key": "2026/2026-03.zip",
16 "size": 13818783,
17 "records": 154,
18 "updatedAt": "2026-04-15T12:11:47.148Z"
19 }
20 ]
21 }

Download Entire Dataset: https://api.sec-api.io/datasets/form-10kt-files.zip?token=YOUR_API_KEY

Downloads the complete Form 10-KT and 10-KT/A archive, covering filings from March 1995 to present, as a single ZIP file. This endpoint requires an API key.

Download Single Container: https://api.sec-api.io/datasets/form-10kt-files/2026/2026-03.zip?token=YOUR_API_KEY

Downloads one monthly container ZIP instead of the full dataset, which is useful for incremental updates based on the updatedAt timestamps returned by the index API. This endpoint requires an API key.

Frequently Asked Questions

What forms does this dataset cover?

The dataset covers Form 10-KT and Form 10-KT/A submissions to EDGAR. Form 10-KT is the transition report filed in lieu of Form 10-K when a domestic registrant changes its fiscal year-end and the resulting transition period exceeds six months; Form 10-KT/A is the amendment variant used to restate, correct, or supplement a prior transition report.

What does one record in this dataset represent?

One record is a single EDGAR submission identified by a single accession number. On disk it is a flat filing folder (named by the dashes-stripped accession number) containing a metadata.json index, the primary 10-KT or 10-KT/A document, and every human-readable exhibit that accompanied the submission.

Who is required to file Form 10-KT?

Domestic operating companies, holding companies, and other commercial issuers subject to Section 13(a) or Section 15(d) of the Exchange Act must file Form 10-KT when they change their fiscal year-end and the transition period is longer than six months, under Rules 13a-10 and 15d-10. Foreign private issuers (who use Form 20-F), registered investment companies (who use N-CSR), and successor registrants adopting a different fiscal year than their predecessor follow separate paths or fall outside this dataset.

What time period does the dataset cover?

The dataset begins with EDGAR's earliest electronic 10-KT records on 1995-03-01 and runs through present. Earlier paper-era transition reports predating the EDGAR phase-in are not included.

What file formats are the records distributed in?

Containers are distributed as monthly ZIP files keyed as YYYY/YYYY-MM.zip. Inside each container, records contain TXT, JSON, HTML, and PDF files — JSON for the metadata.json index, HTML for modern XHTML / inline-XBRL filings, TXT for pre-HTML and plain-text legacy filings, and PDF for exhibits filed in PDF form.

How does this dataset differ from a Form 10-K dataset?

Form 10-K covers a full 12-month fiscal year; Form 10-KT covers only the transition "stub" period (longer than six months) between an old and a new fiscal year-end. The disclosure architecture is otherwise nearly identical under Regulation S-K and S-X, but transition financial statements cover a non-standard span and comparative-period presentation is re-cast around the transition window.

What is excluded from the packaged records?

Binary image and graphic attachments (GRAPHIC-typed entries in documentFormatFiles) and XBRL linkbase and schema files (EX-101.CAL, EX-101.DEF, EX-101.LAB, EX-101.PRE, EX-101.SCH, and the extracted _htm.xml instance) are catalogued in metadata.json with full EDGAR URLs but are not physically bundled. For inline-XBRL filings, the structured facts remain present inside the primary HTML because iXBRL tags live directly in the XHTML body.