[self-promotion] Free Financial Data API For AI Agents And LLMs: Full-text SEC Filings, Earnings Call Transcripts, 13F, Congressional Trades, Short Interest (a Financial Datasets Alternative With A Free Tier)

Disclosure: I run this, it’s Equibles. In May I posted the self-hosted version of the pipeline here. This is the hosted one, which now also covers non-US stocks and is built to be used by AI agents (over MCP) as well as by code (REST).

What’s in it:

• SEC filings, full text (10-K, 10-Q, 8-K, proxies). You can search inside a filing and pull the exact lines, not just metadata
• financial statements and KPIs as reported, each tagged with fiscal period and filing date
• earnings call transcripts, by speaker
• 13F institutional holdings: holders of a stock, or a manager’s full portfolio
• insider transactions (Forms 4 and 5) and congressional trades (House and Senate)
• short interest, daily short volume, fails to deliver, off-exchange volume, a squeeze score, and a model estimate of the next FINRA short interest
• executive compensation, government contracts, buybacks, going-concern flags, IPO filings
• prices for US, European, UK, Hong Kong and Chinese listings (end of day on the free plan), option chains on paid plans
• Fed economic series and CFTC data

How to get at it:

• MCP, for AI agents: https://mcp.equibles.com/mcp (Claude, ChatGPT, Cursor and others)
• REST: https://equibles.com/docs/api
• free plan: 100 requests a day across both, no card. Plus is $9.99 a month for 10,000 a day. Pro adds bulk CSV downloads (short interest, short volume, FTD, buybacks and a few more, one file per year)
• self-host: the open source core (AGPL-3.0) pulls the SEC data itself: https://github.com/daniel3303/Equibles

One detail for the backtesting crowd: short interest bulk rows carry FINRA’s publication date, so you only use numbers that were public at the time.

Why “Financial Datasets alternative” in the title: it’s the name I see most for financial data APIs aimed at AI agents. From their pricing page they don’t have a free tier, and transcripts, congressional trades and short interest aren’t on their list. If I’ve got that wrong, correct me and I’ll fix the post.

So that’s the pitch, a free financial data API for AI agents and the humans who build them. What would make it more useful as a dataset source for you: bulk files for more datasets, point-in-time snapshots, something else?