Company Fundamentals API: Balance Sheet, Cash Flow & Insider Trades for Finance Agents
Prices tell you what the market thinks; fundamentals tell you whether it's right. Here's a fundamentals API shaped for AI agents — statements, dividends, and insider trades in one endpoint — and how it compares to raw SEC EDGAR and the statement-API vendors.
TL;DR
- •Fundamentals = the 'is the business actually healthy' layer: balance sheet, income statement, cash flow, dividends, insider transactions.
- •You can parse it free from SEC EDGAR (XBRL), but normalizing XBRL across filers and periods is real, ongoing engineering — and it's US-only and filing-paced, not query-shaped.
- •API Pick Financials Search returns all five US-company fundamentals behind one POST endpoint — 200 credits/call, only-on-success — pre-shaped for LLM reasoning.
- •Insider transactions (Form 4) and dividend history are the under-used signals: cheap to fetch here, painful to assemble from EDGAR.
- •Pair Financials with SEC Filings (qualitative 10-K/10-Q text) and Markets (price) for a complete grounded view.
Price is opinion; fundamentals are evidence
A market-data call tells your agent what a stock costs. It can't tell you whether the business behind it is growing, drowning in debt, or quietly being sold by its own executives. That's the fundamentals layer — and it's where a research agent goes from "quotes prices" to "forms a view."
Five datasets carry most of the signal: the balance sheet, income statement, cash flow statement, dividend history, and insider transactions.
The EDGAR tax
All of this is public and free in SEC EDGAR. The catch is XBRL. Company-facts JSON is enormous and inconsistent: the same concept is tagged differently across filers, restatements muddy history, and you must map a sprawling taxonomy to a clean schema before anything downstream can use it. Insider data (Form 4) and dividends are separate filing types you parse on their own. Doing this once for one ticker is a weekend; doing it reliably for an agent that queries any company is a maintained pipeline.
Fundamentals as one tool
Financials Search returns balance sheets, income statements, cash flow, dividends, and insider transactions for US public companies behind one POST endpoint — 200 credits per call, only-on-success, pre-shaped for an LLM.
import httpx, os
API, HEADERS = "https://api.apipick.com/v1", {"x-api-key": os.environ["APIPICK_KEY"]}
def fundamentals(query: str):
return httpx.post(f"{API}/search/financials", headers=HEADERS,
json={"query": query}).json()["results"]
fundamentals("Apple latest balance sheet and cash flow")
fundamentals("recent insider transactions at Meta")
fundamentals("Microsoft revenue and net income trend")EDGAR vs. vendor vs. suite
| Parse SEC EDGAR | Statement-API vendor | API Pick Financials | |
|---|---|---|---|
| Cost | Free + engineering | Subscription | 200 credits/call, only on success |
| Setup | XBRL normalization pipeline | Integrate typed endpoints | One endpoint, one shape |
| Coverage | US, filing-organized | Deep, long history | US fundamentals + insiders + dividends |
| Query style | By filing/concept | Typed params | Natural-language |
| Best for | One-off / full control | Fundamentals as the product | Finance agents (one tool of many) |
The complete picture
Fundamentals pair naturally with SEC Filings Search for the qualitative side (risk factors, MD&A, earnings-call language) and Markets Search for price. Together they're the spine of the investment-research agent. One key, JSON throughout, only-on-success. Start free with 100 credits, no card.
Frequently Asked Questions
Can't I just get fundamentals free from SEC EDGAR?
You can — EDGAR's company-facts and XBRL data are free and authoritative. The cost is engineering: XBRL tags drift across filers and periods, restatements complicate history, and you must map thousands of taxonomy concepts to a clean schema before a model (or a chart) can use them. It's also US-only and organized by filing, not by question. For one company occasionally, EDGAR is fine. For an agent querying many companies in natural language, a normalized API saves weeks.
What's covered — and is it US-only?
Balance sheets, income statements, cash flow statements, dividend history, and insider transactions for US-listed public companies. It's US fundamentals; for non-US issuers you'd use a different source. For the filing text itself (10-K risk factors, 10-Q MD&A, earnings calls), use the separate SEC Filings Search endpoint.
Why are insider transactions and dividends worth highlighting?
They're high-signal and annoying to assemble yourself. Insider buying/selling (SEC Form 4) is a watched behavioral signal; dividend history matters for income and quality screens. Both require parsing specific filing types from EDGAR. Here they're first-class fields in the same endpoint as the statements.
How does this compare to financial-statement API vendors?
Dedicated vendors (e.g. financial-statement APIs) give deep, typed endpoints and long history — great if fundamentals are your whole product. API Pick's angle is different: fundamentals as one tool among a suite (markets, SEC, economic, news) behind one key, query-shaped for an agent, billed only-on-success. If you're building a finance agent rather than a fundamentals product, the suite-in-one-shape is the faster path.
Is this investment advice?
No. It's structured public financial data for research and agent grounding. It informs analysis by a qualified person or a supervised agent; it is not advice and must not drive automated trading without human accountability and risk controls.
APIs used in this article
Sarah Choy is the CEO of API Pick. She writes about building production-ready APIs for AI agents and LLM workflows.