he4rt/marketing-extension

He4rt Analytics is a Chrome browser extension that quietly captures the data that X, formerly Twitter, already sends to your browser, and turns it into a structured JSON file. It was built by the He4rt Developers community in Brazil to feed their own backend, called He4rt Hub, with detailed engagement data about a specific account. The README is in Portuguese. The problem it tries to solve is that X does not let community managers export engagement data in bulk. You can see likes and views one tweet at a time, but you cannot list every community member who interacts with you, track replies across posts, or cross reference the people who liked your posts with your own member database. The extension runs in the background while you browse X, intercepts the GraphQL responses the page is already loading, and merges them into one clean export. Internally there are three parts. A script called interceptor.js runs in the main world of the page so it can replace the fetch function and clone every response. A second script, content.js, runs in the isolated extension world and forwards those messages to the background. The background service worker, background.js, filters by the handle you are tracking, deduplicates by tweet id, and consolidates everything. The README explains the split is needed because the X site has a strict content security policy, so the page level patch and the chrome.runtime access have to live in separate worlds and talk through window.postMessage. The endpoints it captures are UserTweets for the timeline of the tracked handle, UserByScreenName for the profile data, Favoriters for the users who liked individual tweets, and TweetDetail for reply threads to the tracked handle. Other GraphQL payloads are saved raw for later use. To install it, you clone the repo, open chrome extensions, turn on developer mode, click Load unpacked, and pick the folder. To use it you click the extension icon, type a handle such as He4rtDevs, click Track, open that profile, and scroll through the tweets. To pull in who liked a tweet, you click the like count and scroll the modal. The popup has an Export JSON button that downloads a file named after the handle and the date. The output groups tracked_account, tweets, community_replies, favoriters_by_tweet, and a summary block with totals and top tweets. The README notes the file is shaped for direct ingestion into the He4rt Hub Laravel backend.