An independent project that turns SEC filings into structured data anyone — or any AI — can use.
OpenCapital takes the 10-K, 10-Q, and 8-K filings every public U.S. company submits to the SEC and turns them into clean, structured pages — one per company, one per metric, one per reportable segment. The same data also powers a Model Context Protocol (MCP) server that connects Claude and ChatGPT directly to the dataset, so you can ask "how is Apple's Services business doing?" and get an answer grounded in the filings instead of a Wikipedia summary.
For a detailed walk-through of where the numbers come from and how we handle restatements, fiscal years, and segment changes, see the methodology page.
OpenCapital is built and operated by Alexey Shishlyannikov, a software engineer who got tired of paying terminal-priced fees to ask basic questions about public companies. The project started as a personal tool for parsing SEC filings, grew into a public website, and now also exposes an MCP server so AI assistants can use the same data.
The team is intentionally small. There is no sales department, no analyst desk, no investor relations layer between the people writing the parsers and the people using the site. If you email support, you will hear back from someone who can actually fix the thing you reported.
All of the public pages on opencapital.sh — every company profile, every metric page, every earnings page — are free and do not require an account. We do not paywall historical data, and we do not sell user data to advertisers or data brokers. Connecting Claude or ChatGPT to the dataset via our MCP server is a paid product; subscription revenue from MCP access is what pays for the filings pipeline that the free pages run on.