User repository on GitHub
Chrome extension: score every X post with one Jev (System One Model) call — worth engaging, buzz, misread, repost/bookmark-worthy, AI-ish
- Forks
- 0
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.
Jev directory
CommunityJev Tools & CLIs
Jev Tweet Radar
A Chrome extension (MV3) that scores every post entering your X timeline with a single Jev call — worth engaging plus selectable tags like spam, buzz, misread and AI-ish, shown as calibrated probabilities in the avatar column, with the goal you pick steering the worth-engaging question.
- Stars
- 1
- Language
- JavaScript
- License
- MIT
- Reading time
- 3 min
- Published by
- Community
- Author
- ddnim
- Last verified
- Oct 2026
- Languages
- JavaScript
- Platforms
- Chrome extensionBrowser
- Capabilities
- Open sourceRequires API key
Highlights
- One POST to api.typesafe.ai/v1/systemone scores each visible post: about 300 input tokens, roughly $0.00001 per judgment, with output free.
- Worth engaging follows a goal — grow followers, learn, excitement or custom text — passed as state.viewer_goal; other tags stay objective.
- Tags show as vertical bars plus a rarity tint in the avatar column: over 75% gold, 60% purple, 45% green; posts are judged within 2000 px of the viewport.
- Filter conditions like spam ≥ 0.7 or worth engaging < 0.3 dim matching posts to 50% opacity, and tags used in conditions are queried even when hidden.
- A 0.6 s long-press bookmarks and likes, then opens a repost, quote or reply window by score; spam or AI-ish above 85% collapses the post.
Watch out
Grounded on the truncated bilingual README and GitHub metadata read on 2026-10-15; the English section breaks off mid-sentence, so the Japanese sections fill in the mechanics. The README itself warns that Jev returns no reasons, that calibrated is not the same as always right, and that post text is sent to TypeSafe; the iOS build needs re-signing every 7 days on a free Apple ID.
Questions about this entry
Short answers drawn from the linked project or page.
A Chrome extension (MV3) for X: every post that scrolls into view gets one Jev call, and the calibrated probabilities for worth engaging plus selectable tags — spam, buzz, flame, ignored, misread, repost-worthy, bookmark-worthy, AI-ish — appear in the avatar column, without a badge row under the post text.
An IntersectionObserver watches article[data-testid=tweet] and extracts text, author, time and media when a post comes within 2000 px ahead of the viewport. background.js sends the post as state plus your selected noul questions to api.typesafe.ai/v1/systemone in one request, retrying 429 and 529 with exponential backoff.
About 300 input tokens, roughly $0.00001 per judgment, with output free; the popup shows today's count and an estimate. A per-minute cap is configurable in the options, and the same post is cached in chrome.storage.local so it is never asked twice.
It changes the worth engaging question to your criteria — grow followers, learn, excitement or custom text — sent as state.viewer_goal. The other tags stay objective, and each goal keeps its own cache so judgments do not mix.
0.6 seconds on the avatar column adds a bookmark and a like, then picks by score: repost-worthy above 50% reposts, above 70% opens the quote window instead, and worth engaging above 70% opens the reply window. Only one window opens, and spam or AI-ish above 85% makes long-press block the account.
Yes. Conditions like spam ≥ 0.7, AI-ish ≥ 0.7 or worth engaging < 0.3 show matching posts at 50% opacity until you hover them, and posts with spam or AI-ish above 85% collapse to the name row until clicked. Tags used in conditions are still queried even if hidden.
There is an iOS Safari Xcode project generated with safari-web-extension-converter; it only works in Safari, and a free Apple ID needs re-signing every 7 days. The README warns that Jev gives no reasons, that judgments are not always right, and that post text is sent to TypeSafe.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
X: An X timeline labeler that flags engagement bait
X: Unclutter: an ad and slop blocker that runs on Jev
X: dmx.to reshapes the timeline in plain language
X: A real-time ad blocker that classifies every DOM element
More Jev resources like this
An open-source BYOK Chrome extension that applies personal, reversible filters to the X timeline: you choose interests, weights and thresholds, Jev scores eligible post text remotely, and every display decision stays on your device.
Jev Tools & CLIs#community#catalog-import#logicrw
Community
mithrilman/your-signalJavaScript
mithrilman/your-signalJavaScriptA local-first Chrome MV3 extension that hides high-confidence pornographic, sexual-innuendo, and solicitation replies on X status pages behind a recoverable placeholder, using the user's own TypeSafe Jev key.
Jev Tools & CLIs#community#catalog-import#logicrw
Community
bugkiwi/elons-jobJavaScript
bugkiwi/elons-jobJavaScriptA bring-your-own-key Chrome extension that asks Jev one slop/not_slop Choice per X or LinkedIn post as it scrolls into view, then blurs and stamps anything at or above a configurable threshold. Scrolling never blocks, and each post is scored once from a local cache.
Jev Tools & CLIs#community#chrome-extension#moderation
Community
davertor/jev-slop-guardJavaScript
davertor/jev-slop-guardJavaScriptA Chrome extension that hides posts you do not want on your X timeline: tap noise bubbles or type rules in plain words, and Jev judges each tweet one-shot before it scrolls into view. Good hide and Shouldn't have hidden answers stay local as an eval set.
Jev Tools & CLIs#community#catalog-import#madewithjev
Community
icpmacdo/signal-and-noiseHTML
icpmacdo/signal-and-noiseHTMLA browser extension that scores X timeline posts in real time with Jev and folds low-signal content while keeping it expandable. Posts are rated 0-100 on information signal, actionability and originality, and without an API key the extension runs in local mock mode.
Jev Tools & CLIs#community#catalog-import#logicrw
A Manifest V3 Chrome extension that sends each LinkedIn feed post's text to TypeSafe's Jev with a rubric of narrow yes/no and 0–2 questions, then blurs only the posts that plain code marks as filter — uncertain and failed checks stay visible, with one-click reveal.
Jev Tools & CLIs#community#catalog-import#madewithjev
From the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.