what is har fr?

HAR Capture Suite is a tool for recording every network request a Chrome tab makes, including requests that Chrome's own built in developer tools sometimes miss, such as traffic inside embedded iframes for captchas or payment forms, or full page navigations during a signup form submission. It is made of two parts working together: a Chrome browser extension that watches network activity using Chrome's debugging protocol, and a separate desktop app built with Electron and React that receives that traffic and displays it live. Once paired together using a security token, the extension streams every request it sees to the desktop app over a local connection on your own computer. The desktop app shows requests in a scrollable, searchable timeline that can handle very large sessions, and lets you click into any request to see its headers, payload, and response body, with JSON responses formatted for readability. You can also inspect WebSocket messages, right click a request to copy it as a curl command or fetch call, and import HAR files you already have. A feature called sticky tab capture means that once a tab starts being recorded, it keeps recording even as you navigate to other websites in that same tab, which is useful for following a full checkout flow that hops between different domains. The tool can also detect captchas on a page. Captured sessions are saved locally in a SQLite database so you can come back to past sessions later, and everything can be exported as a HAR file or a ZIP archive, with an option to automatically hide sensitive information like authorization headers and cookies before exporting. To use it, you install the project's dependencies with npm, build and load the Chrome extension in developer mode, and run the desktop app, then pair the two together using a token shown in the desktop app. After that, you add the websites you want to watch to an allowlist and Chrome will start capturing their traffic automatically. Requires Node.js and Chrome, Chromium, or Edge on Windows, Mac, or Linux.