LinkedIn Farming & Scam Blocker

Two things clog a LinkedIn job search. Engagement farming: the post says it's hiring, but the link is behind "comment INTERESTED and I'll DM you," or a repost, or a follow. The job may not exist. The engagement is the point. Fraud: contact only over WhatsApp or Telegram, a registration or training fee, "direct joining, no interview," data entry at $55 an hour, reshipping and mystery shopper work, an apply link behind a URL shortener. LinkedIn filters neither. This extension does, entirely in your browser. WHAT IT DOES A flagged post or listing collapses to one line naming the rule that fired, with a Show button. Nothing is deleted. A floating counter tells you how many are hidden and reveals them all in one click. 15 rules are on by default, covering both engagement farming (comment-gated links, DM gates, repost bait, follow bait, emoji walls) and scams (WhatsApp/Telegram handles, fees and deposits, no-interview guarantees, reshipping and mystery-shopper genres, URL shorteners, free-email recruiter addresses, body-shop vocabulary, implausible pay for low-skill roles, one poster spamming many listings, the same listing reposted repeatedly). A listing collapses on one scam or farming hit, or on two weak hits. A single weak signal only dims it, because that's where false positives live. A negation guard keeps posts that warn about these scams (quoting the exact phrases) from being hidden themselves. NO RULE LOOKS AT NATIONALITY, COUNTRY, OR NAME Fraud is posted from everywhere, and origin is a weak predictor with a large false-positive rate. Every rule matches behavior: a request to pay, a push to WhatsApp, a promise to skip the interview. The built-in rule learner (see below) refuses to propose a rule based on shared nationality among marked posts, and tells you why. YOU CAN SEE WHAT IT'S DOING Hover a card and press F to mark it. The options page mines your marked cards for phrases that repeat across two or more of them and offers them as new rules, one click each, nothing added automatically. Marked posters can be blocked by profile URL or name. An allowlist protects companies or phrases that must never be hidden. Hit counts per rule make an over-firing rule obvious immediately. Master on/off switch. Per-rule checkboxes. Three modes for flagged items: collapse to a reason, hide completely, or just dim. PRIVACY No network calls, no account, no telemetry, no analytics. Marked cards and settings stay in Chrome's local storage on your machine. Source is public; the privacy claim is one grep away from verified yourself. PERMISSIONS storage (to save your settings and marked cards locally) and content-script access to linkedin.com. Nothing else. Source: https://github.com/Mariano215/linkedin-farming-blocker