Financial data is informational and may be inaccurate. See Disclaimer.

drillr.ai
API keys →

Introduction

Drillr is financial data for the US, China and Japan — companies, filings, financial statements and metrics, prices, earnings, ownership, events, analyst and executive data, and research signals read out of filings, news and earnings calls. It is built in-house from primary sources and served two ways: a REST API for scripts and pipelines, and an MCP server for agents. Same tables, same citations — pick by what is doing the reading.

Ways in

In a hurry? Quickstart is one request, start to finish.

What you get

The data a company analysis needs, in one place and one shape — and the surface keeps growing.

  • Companies and tickers. Profiles, identifier resolution and natural-language company discovery.
  • Filings. Indexed filings with deep links to the originals, and full-text search inside them.
  • Financials. The three statements as reported, plus precomputed valuation and quality metrics.
  • Prices and earnings. Historical bars, the earnings calendar with estimates and actuals, call summaries.
  • Ownership and people. Insider trades, institutional holdings, executive rosters and compensation.
  • Events, analysts and signals. A typed corporate-event timeline, rating changes, and research signals such as industry inflections and enterprise AI adoption.

Where the data comes from

Drillr builds its data in-house, along three lanes.

  • Parsed from filings. Statements, metrics, ownership, executives and filing-sourced events are extracted from the documents each market publishes — SEC EDGAR, the Chinese exchanges’ disclosure platforms, EDINET and TDnet — and every figure keeps a link to the filing it came from.
  • Extracted from news and earnings calls. Event streams, rating changes, call summaries and research signals are drawn from press releases, news coverage and call transcripts, with the source article or excerpt kept alongside.
  • Estimated by Drillr’s own models. Forward-looking figures such as EPS and revenue estimates come from models Drillr runs itself, and are replaced by actuals once disclosed.

Coverage

Companies, filings, the three statements and metrics are available across the United States, China and Japan. The US goes deepest today, adding prices, ownership, events, analyst and executive data; the other markets are filling in behind it. Each endpoint page states the markets, periods and limits it serves.

What makes it different

  • Figures carry their source. Numbers come out of the filing, not a normalised vendor table, and the filing is in the payload — an agent can cite it and a reader can check it.
  • Find before you fetch. Company discovery takes a plain-language description and returns the companies that match, so the ticker does not have to be known up front. Filing search reads inside the filings themselves and returns the passage and the section it sits in, not just the document.
  • Signals you cannot buy elsewhere. Industry inflections, enterprise AI adoption and cross-source news events are conclusions, not pass-through fields — each one extracted and aggregated by Drillr’s research pipeline with its evidence attached.
  • One convention across markets. The same fields and period semantics for the US, China and Japan; only the ticker suffix changes.
  • Built for agents as much as scripts. Every REST endpoint is a GET; the MCP server exposes the same tables as tools, and one credit wallet covers both.

Next steps

  • Quickstart — a key and a first request.
  • REST API — endpoints, parameters and examples.
  • MCP server — connect Claude Code, Codex, Cursor or Claude Desktop.