> ## Documentation Index > Fetch the complete documentation index at: https://docs.agentkey.app/llms.txt > Use this file to discover all available pages before exploring further. # SEC filings and ownership data for AI agents > EDGAR filing search and full text, 10-K sections, XBRL statements, insider trades, 13F and 13D/13G holdings, executive pay and enforcement for AI agents. Everything a US public company must disclose is on EDGAR, but EDGAR is built for humans with a browser: no structured search across forms, no clean way to pull one section of a 10-K, no normalized insider or 13F tables. Through AgentKey your agent queries the filings as data — `find_tools`, `describe_tool`, `execute_tool` — with sec-api's structured datasets, the SEC's own EDGAR endpoints, and Finnhub and AlphaVantage ownership feeds behind the scenes. ## What you can ask * *"Find Apple's latest 10-K and give me the Risk Factors section as plain text."* * *"Which companies mentioned 'generative AI' in an 8-K this year?"* * *"List every Form 4 insider sale at Nvidia in the last 90 days."* * *"What did Berkshire Hathaway hold at the end of last quarter, per its 13F?"* * *"Who has crossed 5% ownership in this small cap recently?"* * *"How much was each named executive at Tesla paid last year?"* * *"Who sits on Boeing's board, and which committees are they on?"* * *"Has the SEC brought enforcement actions against this investment adviser?"* ## What's covered **Filings search and documents** * Search every EDGAR filing from 1993 to present by ticker, CIK, form type, filing date and more; results carry accession number, period of report, primary-document and exhibit URLs * Full-text search across filings and their exhibits (10-K, 10-Q, 8-K, 20-F, S-1 and more) since 2001, narrowable by form type, filer, SIC code and date * Extract one section — Risk Factors, MD\&A, Business — from a 10-K, 10-Q or 8-K as plain text or the original HTML * Structured 8-K material events with item numbers and event descriptions * Company EDGAR profile and filing history by CIK; CIK lookup by name or ticker; cross-mapping of CIK, ticker, CUSIP and name; the EDGAR entity master **XBRL financial data** * XBRL from a 10-K or 10-Q converted into normalized JSON statements — income statement, balance sheet, cash flow and disclosures * Every reported value of one XBRL concept for a company across its filings; one concept across every filer for a calendar period, for screening **Insider and institutional ownership** * Insider transactions from Forms 3, 4 and 5 — one route covers US, UK, Canada, Australia, India and major EU markets; another covers Form 4 for US-traded companies * Insider sentiment for US companies; Form 144 intent-to-sell notices * Form 13F institutional holdings and cover pages; institutional holders of a US-traded company * Schedule 13D / 13G beneficial-ownership filings at the 5% threshold * Fund portfolio holdings (N-PORT), fund census (N-CEN) and proxy-voting records (N-PX) * Congressional trading activity for a company **Governance and compensation** * Executive compensation — salary, bonus, stock and option awards, non-equity incentive, total — by search or by ticker; director compensation where reported * Directors and board members with roles, committees, independence status and term dates * Auditor and audit-fee disclosures; Exhibit 21 subsidiaries with jurisdictions; public float and share count **Offerings and registrations** * S-1 / 424B4 prospectuses for IPOs and follow-ons; Form D private placements; Form C crowdfunding; Regulation A offerings and their 1-A, 1-K and 1-Z forms * Investment adviser registrations (Form ADV firms, individuals, brochures and ownership schedules); broker-dealer FOCUS reports **Enforcement** * SEC enforcement actions, litigation releases, administrative proceedings, accounting-and-auditing enforcement releases and SRO rule filings ## How to use it Pass the request in plain language and let the router pick the tool: ```text theme={null} find_tools(q="risk factors section of Apple's latest 10-K") find_tools(q="伯克希尔最新的 13F 持仓") find_tools(prefix="finance/secapi") # structured EDGAR datasets, section extractor, XBRL find_tools(prefix="finance/sec") # EDGAR submissions, XBRL concepts and frames, full text find_tools(prefix="finance/finnhub") # filings list, insider transactions and sentiment describe_tool("") execute_tool("", { ...params }) ``` Name the form ("10-K", "13F", "Form D") and the router ranks the matching dataset first. Always `describe_tool` before executing — several datasets take a Lucene query string, and field names and per-call costs differ between providers. > [!NOTE] > > **Building with code?** The same capability is available as fixed-contract > [Scenario Methods](/api-reference/scenario/overview) with automatic provider > failover: > [`finance.filings.search`](/api-reference/scenario/methods/finance-filings-search), > [`finance.filings.get`](/api-reference/scenario/methods/finance-filings-get), > [`finance.ownership`](/api-reference/scenario/methods/finance-ownership). ## FAQ > **Do I need a CIK?** > > Not to start. Look a company up by name or ticker to get its CIK, or use > the identifier mapping to go from ticker, CUSIP or name to the full > identifier set. Many datasets also accept a ticker directly. > **How far back do filings go?** > > Filing metadata search covers 1993 to present; full-text search covers > filings and exhibits since 2001. XBRL-based data exists only for filings > that carry XBRL. > **Is it only US companies?** > > EDGAR includes foreign private issuers (20-F filers) and every registered > fund and adviser. Insider-transaction coverage on one route extends to the > UK, Canada, Australia, India and major EU markets. This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.