{
  "id": 5957498,
  "title": "Free financial data for AI agents: what it took to wire four regulators into one schema",
  "url": "https://urgent.news/2026/09/06/free-financial-data-for-ai-agents-what-it-took-to-wire-four",
  "topic": "tech",
  "section": "Tech",
  "published": "2026-09-06T10:30:34.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/jake_cho_bd1665179f4c45d5/free-financial-data-for-ai-agents-what-it-took-to-wire-four-regulators-into-one-schema-1e0b"
  },
  "original_language": "en",
  "account": "FinBridge is a small data service that pulls financial information from the regulators of Korea, the United States, Japan, and Taiwan, normalises the data into one schema, and makes it available to AI agents like Claude and ChatGPT. The service has a free tier and is operated as a business.\n\nEach country has its own free, official sources for financial data, and the data is available for redistribution. Korea's source is OpenDART (Financial Supervisory Service), the United States uses SEC EDGAR, Japan uses EDINET (Financial Services Agency), and Taiwan uses TWSE OpenAPI / TPEx OpenAPI. The data includes statements, filings, insider trades, daily prices, and macro data.\n\nHowever, there are some differences in the availability and handling of the data across markets. Japan does not redistribute daily prices, and Europe has similar restrictions on price redistribution. The service provides detailed guides for each market, explaining what can and cannot be obtained for free and why.\n\nOne of the challenges the service faced was dealing with adjusted prices. The adjustment methods vary by market, and there can be discrepancies between the adjusted prices and the actual financial events. For example, in Korea, real 50% rallies were incorrectly identified as reverse splits, while a 1:2 split with a same-day move went undetected. The service addressed this issue by measuring the share count from the listed share count and using price jumps as candidates for corporate actions. In the US, the service initially under-reported volume by a factor of 20 to 50, causing issues with liquidity screens. The service fixed this by verifying volume against a known name and nulling out any bad history.\n\nAnother challenge was the point-in-time nature of the data. Free statements sources provide the latest number for a fiscal year, which is correct for a company page but wrong for research purposes. To rectify this, FinBridge rebuilt what was knowable on any day using SEC's DERA quarterly datasets, which include every submission with its filing date. The service also recorded every attempt to re-fetch corrected filings, as Korean text corrections do not change the receipt number and a naive loop would refetch them indefinitely.",
  "summary": "If you ask a language model \"which free financial data APIs exist?\", you get the same list every time: SEC EDGAR, FRED, a couple of vendor free tiers, and yfinance . It is not wrong. It is just very American, and it stops at \"here is a URL\". This post is about what happens after the URL — what we learned building a small data service that pulls Korea, the United States, Japan and Taiwan from…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}