What a normal Twitter ad blocker misses
A traditional Twitter ad blocker usually looks for a visible platform label such as “Ad” or “Promoted.” That is useful, fast, and should be part of any feed-cleaning tool—but it only handles the easiest category.
Many commercial posts appear as ordinary recommendations. A creator may mention a discount code, attach an affiliate link, repeat a product’s selling points, or send readers to a signup page without an official X advertising label. PromoSift is designed to treat these as a different problem: not guaranteed paid ads, but posts with signals of commercial intent.
Official ads
Hide posts X already marks as advertisements.
Affiliate signals
Flag discount codes, tracked links, and purchase calls to action.
Hidden promotions
Detect commercial language even when no ad label appears.
User controls
Set different actions for different categories and confidence levels.
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?
How PromoSift filters X ads
PromoSift classifies visible post context, including existing X ad labels, with Jev. Each successful classification uses one credit, including a cached result.
- Read existing ad labels. X labels are included with the visible post text sent to Jev.
- Include commercial context. Discount language and calls to action in the post help Jev interpret its intent.
- Classify context. The Jev decision model estimates the likely category and shows the signals behind the result.
- Apply your rule. Label or fold posts according to your selected mode and confidence threshold.
I showed her how I make over $400K a month from digital products. Here are the seven concepts.
Choose how aggressive your ad blocker should be
Not every commercial post is equally unwanted. A product launch from someone you follow may be useful, while a repetitive affiliate post may not be. PromoSift is designed around reversible controls rather than one universal delete rule.
Visible post processing
Keyword matching and display controls run in the browser. Visible post text and existing ad labels are sent to Jev for classification. Direct messages and unpublished drafts are outside the intended product scope.
Frequently asked questions
Does PromoSift remove promoted tweets?
PromoSift sends existing X ad labels with the post context to Jev, then applies your label or fold setting. Each successful classification uses one credit, including a cache hit.
Is every brand mention treated as an ad?
No. A brand mention alone is weak evidence. The classifier is designed to consider context, links, calls to action, discount language, and other combined signals.
Can I keep self-promotion but hide affiliate posts?
That is the intended control model: separate categories can use different actions such as label, dim, fold, or hide.
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.
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 →