X Shares Its Feed Ranking Code and Lets Users Check If Their Posts Were Affected

X announced on August 13, 2026 that it is open-sourcing its 'For You' ranking algorithm and adding a transparency tool that lets users see if their account or posts have been impacted by its ranking systems TechCrunch.

The 'For You' feed is the main timeline X shows users, and the ranking algorithm is the software that decides which posts appear there and in what order. X published the source code for this system on GitHub at github.com/xai-org/x-algorithm under the Apache v2 license, which lets anyone view, use, and modify the code freely TechCrunch. The new version is roughly 10 to 15 times larger than the previous release and includes details about the parameters used to weight different signals — meaning the factors the system uses to decide how high or low a post should appear.

VP of Product Keith Coleman said the release includes the core ranking code that pulls and ranks posts for any given user and assembles the feed. Some parts of the system, like the ranker and the score (a number the system calculates for every post to decide how visible it should be), can be run outside the company TechCrunch. Before launch, X showed the code to outside researchers who study recommendation systems, and they were able to get the scoring system running on their own computers.

Developers can submit pull requests through the GitHub repository. A pull request is a proposed change to the code, and X engineers will consider incorporating those changes into the algorithm TechCrunch.

The release does exclude some systems. For example, parts that use Grok, X's AI model, to predict whether a post might break a rule are not included. X says this prevents bad actors from studying the code to work around the rules and flood the network with spam TechCrunch.

On the user side, X is rolling out a new tool on an 'Under the Hood' page in the app's settings. Users who have posted ten or more times in the past month can download a file (in a format called JSON, which is a way of organizing data so both programs and people can read it) showing any labels applied to their account or posts over the past calendar month TechCrunch. The tool will first be available as a pilot to a test group of accounts at least a year old before rolling out more broadly.

X's Help Center page 'Debunking X myths' states that X does not shadow ban and that it ranks posts to create a more relevant experience, with users always able to see posts from people they follow X Help Center.

Researchers have previously said the version of the algorithm X published earlier does not offer the kind of transparency that would actually be useful Engadget. That assessment was published in February 2026, before this expanded release.

The question now is whether this much larger codebase, combined with the transparency tool, changes that earlier view. In my view, the inclusion of the weighting parameters and the ability for outside researchers to run the ranking and scoring system on their own machines do represent a real step forward. Whether that step is large enough to satisfy critics who want full visibility into how X classifies and moderates content is a separate matter. The fact that X excluded the Grok-based rule-violation systems draws a clear line: the company is willing to show how it ranks posts, but not how it enforces its rules.

For developers and researchers, the open-source release and the ability to propose code changes through pull requests open a relatively unusual channel. Platforms of X's scale rarely invite outside code contributions to the systems that decide what users see. The practical impact will depend on whether X engineers actually adopt community-submitted changes or simply treat the repository as a read-only window into how the algorithm works. The GitHub repository serves as the canonical source for the code, while the announcement and feature details come from the August 13 reporting TechCrunch.

The requirement of ten posts per month to access the transparency tool means casual or new accounts are excluded from the pilot. Combined with the one-year account age threshold, this limits the initial rollout to a relatively established user base. Whether that is a deliberate choice to reduce noise or simply a scaling constraint is not stated.

What this enables, at minimum, is a structured way for technically inclined users to check the labels applied to their own account. For the broader research community, it provides a more detailed view of how X weights different ranking signals than was available before. Whether that amounts to useful transparency is still debated, and the February researcher assessments will likely be revisited as the expanded codebase is examined in the coming weeks.