Remix.run Logo
Show HN: Grammarly for tweet reach – 36 rules from X's source(github.com)
4 points by aytuncyildizli 11 hours ago | 2 comments
aytuncyildizli 11 hours ago | parent | next [-]

I built this because I kept mass-deleting draft tweets that got no engagement. Wanted to know before posting whether something would actually reach people.

It scores against 36 rules derived from X's open-sourced algorithm (twitter/the-algorithm on GitHub). The part I find most useful is the what-if scenarios — remove a link (+52%), add an image (+38%), post at peak time (+25%).

Everything runs client-side in the browser. No account needed. AI features (slop detection, auto-optimize) are optional and use your own API key.

Happy to answer questions about the scoring model.

aytuncyildizli 10 hours ago | parent | prev [-]

[dead]