Clean the commercial noise out of your X feed.

PromoSift helps your timeline feel less like a storefront—without hiding useful posts you still want to see. Powered by Jev, TypeSafe AI’s newly released System One decision model.

Powered by JevCleaner For You feedCategory-level controlsReversible filteringEnglish + Chinese
J
Jev-powered classification

Visible post state in. Typed promotion decisions, probabilities, and confidence out.

About Jev ↗

A feed cleaner is more than an ad blocker

People install feed-cleaning extensions because the problem is broader than one advertising label. Commercial noise can include official ads, affiliate recommendations, repeated product launches, giveaway funnels, newsletter pushes, and self-promotion. Some of it is useful; some of it overwhelms the timeline.

PromoSift focuses on the commercial layer of feed clutter. It is designed to identify likely promotional patterns and let each user decide which categories deserve attention.

01

Quiet mode

Reduce the prominence of likely commercial posts.

02

Category rules

Choose different actions for ads, affiliate posts, and self-promotion.

03

Reason labels

See why the extension classified a post as promotional.

04

Easy undo

Reverse a fold or hide action without losing the post forever.

J
Why PromoSift uses Jev

Traditional blockers match labels and selectors. Jev handles the bounded semantic decision: what kind of promotion is this, how strong is the commercial intent, and how confident is the result?

Official announcement ↗

Turn one noisy feed into a set of controllable categories

Instead of treating every unwanted post the same, PromoSift separates content into practical groups: labeled advertisement, disclosed sponsorship, likely affiliate promotion, self-promotion, general commercial intent, and ordinary content.

That classification makes a more nuanced feed possible. You might hide official ads, fold affiliate posts, label self-promotion, and leave ordinary product discussions untouched.

Lead-gen · 89%Jev decision breakdown
A
Alex Creator @creator_lab

I showed her how I make over $400K a month from digital products. Here are the seven concepts.

Ad probability 89%“make money” 98%AI estimate ~30%

Four ways to clean without losing context

Label adds an explanation while keeping the post visible. Dim reduces visual dominance. Fold collapses the post behind a summary. Hide removes a selected category with an undo path.

The default experience should be conservative. Uncertain content is better labeled than silently removed.

Feed control without reading private conversations

A feed cleaner should only request access needed for its purpose. PromoSift is designed for visible timeline content, not direct messages, private drafts, passwords, or unrelated browsing history.

Frequently asked questions

Can PromoSift remove suggested posts too?

The first product focus is commercial content. Broader recommendation controls may be added later without diluting the core purpose.

Will it clean both For You and Following?

PromoSift is designed for loaded feed posts on the main X timeline, with coverage expanded as X page layouts change.

Can I use a lighter mode that only adds labels?

Yes. Label-only mode is central to the product because many users want awareness without automatic removal.

What is Jev in PromoSift?

Jev is TypeSafe AI’s System One decision model. PromoSift uses it to classify visible post context into typed promotion categories and confidence values that can be mapped to label, dim, fold, or hide actions.

Passwordless access

Detect what basic blockers miss.

Enter your email, verify the six-digit code, and receive a one-time grant of 500 free Jev decisions that never expire. Click daily check-in to claim 50 more each Beijing day (UTC+8); missed days cannot be claimed later.

Log in to PromoSift →