Secedgar Get Filing
secedgar_get_filingFetch a specific filing's metadata and document content by accession number. Returns the primary document as readable text. Use offset/next_offset for multi-page access to large filings (10-K, S-1 can exceed 1M chars): pass the next_offset from a truncated response as offset, without section or find, to read the next page. Use section to jump directly to a heading (e.g. 'risk factors', 'item 7') without needing an offset, or find to open the page at a phrase in the text (e.g. 'net revenue retention') and list where it occurs.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"accession_number": "<accession_number>"
},
"name": "secedgar_get_filing"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| cik | No | Filer CIK, digits only (resolve via secedgar_company_search if you have a ticker or name). Optional but recommended — speeds up archive lookup. If omitted, the filer is looked up from the accession number through EDGAR full-text search (2001 onward) and the accession number's own prefix, which for many filings is a filing agent's CIK, and when none of those directories serves the filing, through the filers its index page names. | |
| find | No | Open the page at a phrase in the document's text (e.g. 'net revenue retention') and list where it occurs. Matched as a substring, ignoring case, whitespace runs, and quote style, so a phrase broken across a line matches; no patterns or whole-word matching. The page opens at the first occurrence at or after offset — at its line's start when that is at most 200 characters earlier, else at the first word starting within those 200 — and occurrences lists up to 20 from there with context, occurrences_total counting the whole document. Blank reads as omitted. With find, offset is where the search starts, so leave find out when paging on with next_offset; section takes precedence over find. | |
| offset | No | Character offset into the extracted document text. Pass next_offset from a truncated response to continue reading the next page, and leave section and find out: section takes precedence over offset, and with find, offset is where the search starts. Default 0 reads from the beginning. | |
| section | No | Jump to a named section, matched against every detected heading and the document's own lines (e.g. 'risk factors', 'item 7', 'item 2.02', 'certain relationships'). The first match wins, in this order: a heading equal to the value, or whose title after its Item/Part marker equals it; the first Item or Part heading containing it; a document line equal to it — a subheading the outline omits ('Liquidity and Capital Resources') — unless the line labels a table row; the first other heading containing it; a table-row label equal to it, else a document line starting with it. In a proxy statement (DEF 14A and its variants), whose ITEM headings are ballot proposals, the first heading of any kind containing the value comes before both document-line steps. Within each heading step a body heading wins over a table-of-contents row, which lands only when nothing else at that step matches; a table-of-contents row that links to its section lands on that section. A line landing says so in the notice; table-of-contents rows and wrapped cross-references are never line landings. A leading 20-F sub-item label on a heading or line ('3D.', 'B.') is ignored, so 'risk factors' reaches '3D. RISK FACTORS'; the value itself is matched as given. A value ending in a number matches only that number: 'item 1' reaches Item 1 and Item 1A, never Items 10–16, and 'part i' never reaches Part II. Every step ignores case, whitespace, and quote style, so a heading copied from the outline resolves whether it carries the filing's non-breaking spaces and curly quotes or plain ones; the steps comparing a whole heading or line (equal to, starting with) also ignore the punctuation between an Item or Part marker and its title, so 'Item 1A Risk Factors' reaches 'Item 1A. – Risk Factors'. For a phrase in the text that is not a heading, use find. Takes precedence over find and offset, so leave it out when paging on with next_offset. On a miss, the error message includes the detected outline so you can pick the correct heading. | |
| document | No | Specific document filename within the filing (e.g., "ex-21.htm" for subsidiaries list). Omitted or blank: the primary document. Available documents are listed in the response metadata under documents, and a primary_document value is accepted too; entries marked binary hold no text and are rejected. | |
| include_xbrl | No | Include XBRL viewer artifacts and machine-readable taxonomy files (R*.htm fragments, *_cal/_def/_lab/_pre.xml linkbases, *_htm.xml inline instance, *.xsd schemas, MetaLinks.json, FilingSummary.xml, Show.js, report.css, *-xbrl.zip, Financial_Report.xlsx, EX-101.* technical exhibits) under documents.xbrl. Off by default — these dominate filing indexes (~100 entries on a typical 10-K) and are rarely relevant when reading filing content. | |
| content_limit | No | Maximum characters of document text to return per page. 10-K filings can exceed 500,000 characters; S-1/A can exceed 1,000,000. Default 50,000 captures ~12,000 words (typically business overview, risk factors, and MD&A). Increase to 200,000 for full financial statements, or decrease for quick summaries. Page on with offset; section and find open a page at a heading or a phrase. Capped at 200000: a larger value, up to 1000000, returns what 200000 returns. | |
| accession_number | Yes | Filing accession number in either format: "0000320193-23-000106" (dashes) or "000032019323000106" (no dashes). Obtained from secedgar_company_search or secedgar_search_filings results. |
Output Schema
| Name | Required | Description | Default |
|---|---|---|---|
| cap | No | The `content_limit` that was applied. | |
| cik | No | Filing entity CIK, zero-padded to 10 digits. | |
| form | No | Form type (e.g., "10-K"), from the submissions feed or the filing's SEC header. Absent only when neither has it. | |
| error | No | Present when the call failed. Absent on success. | |
| shown | No | Characters of document text returned on this page. | |
| notice | No | How to read the next page or list more occurrences of find, which file was read when the archive does not serve the indexed primary, when section landed on a document line rather than a detected heading, and how many headings were detected when the outline lists fewer. | |
| content | No | Document text content for this page window. | |
| outline | No | Up to 50 detected headings on a truncated first page (offset=0, no section or find). Past 50: Item and Part headings first (bare TOC copies of titled ones dropped), then the earliest; the notice gives the count. Pass text as section (a linked TOC row lands on its section) or offset as offset. | |
| documents | No | Filing documents by category. Any name is a valid document input except entries with binary: true, which fail with binary_document; scans can outnumber readable documents. | |
| truncated | No | True when the document is longer than `content_limit` allowed through. | |
| filing_url | No | Direct URL to the filing on SEC.gov. | |
| filing_date | No | Date the filing was submitted (YYYY-MM-DD). Absent under the same conditions as form. | |
| next_offset | No | Offset of the next page, to pass as offset without section or find; present while content_truncated is true. | |
| occurrences | No | With find: up to 20 occurrences, from the one the page opens at. | |
| company_name | No | Filing entity name. Absent if the CIK did not resolve to a known entity. | |
| period_ending | No | Period of report (YYYY-MM-DD), from the same source as form. Absent for forms without one (S-8, Form 4, proxy statements) or when neither source has it. | |
| accession_number | No | Filing accession number, normalized to dash format. | |
| primary_document | No | Primary document as read: an XML form's SEC rendering (xsl…/form4.xml) up to 600 KB raw; a 2000–2001 name SEC does not serve, as <accession>-<name>, else the full submission <accession>.txt. | |
| content_truncated | No | True if content was truncated at content_limit. | |
| occurrences_total | No | With find: occurrences in the whole document. | |
| requested_document | No | Filename requested via document. Present only when it differs from primary_document. | |
| content_total_length | No | Full document length in characters. |