{"id":1682,"date":"2026-08-14T23:03:04","date_gmt":"2026-08-14T23:03:04","guid":{"rendered":"https:\/\/feedsta.ai\/blog\/x-open-sources-ranking-algorithm-shadowban-tool\/"},"modified":"2026-08-14T23:03:05","modified_gmt":"2026-08-14T23:03:05","slug":"x-open-sources-ranking-algorithm-shadowban-tool","status":"publish","type":"post","link":"https:\/\/feedsta.ai\/blog\/x-open-sources-ranking-algorithm-shadowban-tool\/","title":{"rendered":"X open sources its ranking algorithm and adds a shadowban transparency tool"},"content":{"rendered":"<p>X has released the source code for its For You feed and core ranking engine on GitHub under the Apache 2.0 license, and is rolling out an in-app transparency tool that lets users check whether their posts have been affected by ranking filters. The release expands a previous open source effort and makes the codebase roughly 10 to 15 times larger than before, including the model configuration, filters, and the systems that weight different ranking signals.<\/p>\n<h2>What X is actually releasing<\/h2>\n<p>The repository contains the code that pulls posts, ranks them for any given user, and assembles the feed that appears by default when the app opens. It includes the parameters used to weight different signals, which determine which posts get surfaced. Parts of the system, such as the ranker and the scoring model, can be run by developers outside the company.<\/p>\n<p>The company previewed the codebase to external researchers familiar with recommendation systems before launch. Those researchers were able to train and run X&#8217;s Phoenix scoring system using the open source code, a step X described as a milestone for its transparency work.<\/p>\n<h2>How users can check if they have been shadowbanned<\/h2>\n<p>A new &#8220;Under the Hood&#8221; page in the app&#8217;s settings gives users who have posted 10 or more times in the past month the ability to download an aggregate statistics file in JSON format. The file shows whether any labels have been applied to their account or individual posts over the previous calendar month.<\/p>\n<p>Non-technical users can drop the file into a large language model of their choice, point the model at X&#8217;s GitHub repository, and ask for an interpretation of the data. X said the tool will first roll out as a pilot to a test group of accounts that are at least a year old before expanding more broadly.<\/p>\n<h2>What is not in the release<\/h2>\n<p>Some systems are excluded, including the parts that use Grok to predict whether a post could violate a rule. X said keeping those components closed is meant to stop bad actors from reverse engineering the rules to flood the platform with spam.<\/p>\n<h2>Community contributions and outside scrutiny<\/h2>\n<p>Developers will be able to submit pull requests to the GitHub repository. X engineers will review submissions and may incorporate improvements into the live algorithm. The company framed the move as part of a broader effort to address longstanding concerns about how its ranking system shapes political conversation, elections, and the spread of misinformation.<\/p>\n<p>X&#8217;s stated goal is to let the public audit whether posts are distributed fairly, critique the system, and help improve it. The platform has faced accusations of shadowbanning from political figures for years, including congressional complaints from Republicans during the company&#8217;s earlier ownership.<\/p>\n<h2>FAQ<\/h2>\n<h3>What did X open source?<\/h3>\n<p>X published the source code for its For You timeline and core ranking engine on GitHub under the Apache 2.0 license, including model configuration, filters, and the parameters that weight ranking signals.<\/p>\n<h3>How can users tell if they have been shadowbanned?<\/h3>\n<p>Users with at least 10 posts in the past month can open the new &#8220;Under the Hood&#8221; page in settings and download a JSON file showing any labels applied to their account or posts over the previous calendar month.<\/p>\n<h3>What parts of the ranking system are not included?<\/h3>\n<p>Components that use Grok to predict rule-violating posts are kept closed to prevent spammers from exploiting that knowledge.<\/p>\n<p><script type=\"application\/ld+json\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"headline\":\"X open sources its ranking algorithm and adds a shadowban transparency tool\",\"description\":\"X released its For You ranking code on GitHub under Apache 2.0 and added an in-app tool that shows users if posts have been filtered.\",\"datePublished\":\"2026-08-14T23:01:49.303Z\",\"publisher\":{\"@type\":\"Organization\",\"name\":\"Feedsta\"}},{\"@type\":\"FAQPage\",\"mainEntity\":[{\"@type\":\"Question\",\"name\":\"What did X open source?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"X published the source code for its For You timeline and core ranking engine on GitHub under the Apache 2.0 license, including model configuration, filters, and the parameters that weight ranking signals.\"}},{\"@type\":\"Question\",\"name\":\"How can users tell if they have been shadowbanned?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"Users with at least 10 posts in the past month can open the new \\\"Under the Hood\\\" page in settings and download a JSON file showing any labels applied to their account or posts over the previous calendar month.\"}},{\"@type\":\"Question\",\"name\":\"What parts of the ranking system are not included?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"Components that use Grok to predict rule-violating posts are kept closed to prevent spammers from exploiting that knowledge.\"}}]}]}<\/script><\/p>\n<hr style=\"margin:2.5em 0 1em;opacity:.35\" \/>\n<p style=\"font-size:.85em;opacity:.7\">This article summarizes reporting from <a href=\"https:\/\/techcrunch.com\/2026\/08\/13\/x-open-sources-its-ranking-algorithm-letting-users-see-if-theyve-been-shadowbanned\/\" target=\"_blank\" rel=\"nofollow noopener\">techcrunch.com<\/a>.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>X is releasing its For You ranking code on GitHub under Apache 2.0 and adding a settings tool that shows whether posts have been filtered.<\/p>\n","protected":false},"author":1,"featured_media":1681,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"rank_math_title":"","rank_math_description":"","rank_math_focus_keyword":"","rank_math_canonical_url":"","rank_math_facebook_title":"","rank_math_facebook_description":"","rank_math_twitter_title":"","rank_math_twitter_description":"","rank_math_robots":[],"footnotes":""},"categories":[1],"tags":[],"class_list":["post-1682","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai-news"],"_links":{"self":[{"href":"https:\/\/feedsta.ai\/blog\/wp-json\/wp\/v2\/posts\/1682","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/feedsta.ai\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/feedsta.ai\/blog\/wp-json\/wp\/v2\/types\/post"}],"replies":[{"embeddable":true,"href":"https:\/\/feedsta.ai\/blog\/wp-json\/wp\/v2\/comments?post=1682"}],"version-history":[{"count":1,"href":"https:\/\/feedsta.ai\/blog\/wp-json\/wp\/v2\/posts\/1682\/revisions"}],"predecessor-version":[{"id":1683,"href":"https:\/\/feedsta.ai\/blog\/wp-json\/wp\/v2\/posts\/1682\/revisions\/1683"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/feedsta.ai\/blog\/wp-json\/wp\/v2\/media\/1681"}],"wp:attachment":[{"href":"https:\/\/feedsta.ai\/blog\/wp-json\/wp\/v2\/media?parent=1682"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/feedsta.ai\/blog\/wp-json\/wp\/v2\/categories?post=1682"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/feedsta.ai\/blog\/wp-json\/wp\/v2\/tags?post=1682"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}