jevbooks

← All projects

lkclean

stefw/lkclean

Chrome extension that filters LinkedIn feed noise with Jev and explains each verdict.

What it is

A Manifest V3 TypeScript Chrome extension for LinkedIn users that collapses engagement bait, self-promotion, sponsored and off-topic posts into a one-line bar, and annotates kept posts with their scores.

How it uses Jev

Each post's author and text form a state { user_interests, post } sent to POST https://api.typesafe.ai/v1/systemone with typed questions: engagement_bait, self_promo, matches_interests (noul), top_interest, blocked_topic (choice). decide() in src/background.ts turns the returned probabilities into a verdict using user thresholds.

Primitives:noulchoice

Technique worth stealing

Ask several typed questions in one parallel request, then convert probabilities to verdicts with plain threshold code instead of parsing free text.

Try it

Download lkclean-vX.Y.Z.zip from the latest release, unzip, then chrome://extensions → Developer mode → Load unpacked.

View on GitHub

judged by Jevjev-1.13.0

Evidence

Each line is one question put to Jev about the README. ≥ 0.60 reads as yes, ≤ 0.40 as no; in between Jev is not making a call.

  • Jev-centricyes0.94
  • Shows a System One patternno0.28
  • Handles uncertaintyno0.07
  • Measuredno0.05
  • Runnableyes0.98
  • Worth recommendingno0.38
  • Model replicano0.03
  • Problem scopescore on a 0–2 scale1.02
  • About Jevyes0.98

Signals by Jev jev-1.13.0, card written by DeepSeek V4.1 Flash from the README on 20 Sept 2026.