SEC EDGAR

What it is

The filing system itself — the place every other product in this category reads from. Three surfaces matter to somebody building rather than browsing.

The JSON APIs on data.sec.gov. A submissions endpoint per company returning its filing history, and three XBRL endpoints returning tagged financial facts by company, by concept or across every filer for one period. No key, no account, no registration.

Full-text search. Every filing from 2001 onward, searchable by phrase, form type and date, with a public JSON backend behind the UI that the SEC does not document as a product.

Quarterly flat files. The research sets are how you get insider and holdings data in bulk without writing a parser: Insider Transactions from the first quarter of 2006, flattening Forms 3, 4 and 5, and Form 13F from the third quarter of 2013.

Pricing

Free, and structurally so. There is no plan, no key, no commercial tier and no redistribution licence to negotiate, because these are public records a filing requirement puts there. That is a different kind of free from a vendor's free tier: nobody can withdraw it to sell you an upgrade, and no term of service narrows what you may do with what you downloaded.

What replaces the price is a rate limit. Ten requests a second is the published ceiling, a declared user agent naming you and a contact address is required on every request, and the SEC asks for efficient scripting and reserves the right to throttle. Ignore that and the cost is access, which is the only bill this product can send.

Data & coverage

Everything filed on EDGAR: Forms 3, 4 and 5 for insider transactions, 13F-HR for institutional holdings, Schedules 13D and 13G for large stakes, Form D for private offerings, N-PORT for fund constituents, and the whole 10-K, 10-Q and 8-K series underneath. US registrants and foreign private issuers filing into the US market.

Latency is not the limit people expect. The SEC gives a typical processing delay of under a second for the submissions API and under a minute for the XBRL APIs, so a Form 4 is machine-readable essentially when it is accepted. Everything downstream — every tracker in this category — is slower than this by construction.

The gap worth committing to memory is which endpoint holds what. The XBRL APIs are financial statement facts only. Insider and 13F data live in the raw filings, so you either parse the filing XML, use full-text search to find it, or take the quarterly flat file and accept that it lands a quarter late.

Integrations

Plain REST over HTTPS returning JSON, an RSS feed of recent filings, and nightly bulk archives of the submissions and company-facts datasets. No websocket, no webhooks, no spreadsheet add-in, and no official MCP server — SEC EDGAR MCP is a third-party project over these same endpoints, as are several others.

Identifiers are the practical friction. EDGAR is organised by CIK, not by ticker, and the mapping file between them is a separate download that a reader is expected to join for themselves.

Limitations

• No congressional data. STOCK Act reports go to the House Clerk and the Senate, not the SEC. Lobbying, short interest and government contracts are three more bodies again.
• The XBRL APIs do not carry Form 4 or 13F. The single most common wrong assumption about this product.
• Nothing is normalised across filers. A 13F is what the manager typed. Names, CUSIPs and share classes disagree between filings, and reconciling them is the work every paid vendor in this category is actually charging for.
• Full-text search starts in 2001 and the flat files start later again — 2006 for insider forms, 2013 for 13F. Older filings exist in the archives as documents, not as data.
• No service commitment. No SLA, no support channel, no notice before a page moves. The APIs are published for public use, not sold.
• Ten requests a second is a hard planning constraint. Anything that wants the whole corpus needs the bulk files rather than a loop over endpoints.

Alternatives

sec-api.io is the commercial version of the same data — filing search, XBRL-to-JSON, section extraction and a real-time stream from $55 a month — and the reason to pay is the parsing, not the access. OpenInsider and Insider Screener turn Form 4 into something filterable; 13F.info and WhaleWisdom do the same for 13F. For financial statements as a maintained API rather than raw XBRL, Financial Modeling Prep and Intrinio sit in fundamental data and research platforms.