Is this just another AI detector?

No. It judges whether writing was worth attention using nine slop signs plus voice and usefulness counter-signs. The AI-drafted likelihood can only soften human-seeming scores and never flags a post alone.

What stops it from hiding good writing?

It shows scores out of 100 with spider-chart reasons, counts engagement for benefit of the doubt, and lets No, Maybe, or Probably votes override the model for you. Stiff formal prose and humor can still confuse it.

What data leaves my browser when scoring?

Post text goes to the Slop Mop server and on to Jev. The server keeps only text hashes, LinkedIn ids, scores, counts, and tallies, skips author names, and keeps check logs for 90 days for the admin dashboard.

Can I run it without using your server?

Yes, both halves are MIT. Build the MV3 extension against your server URL, deploy the Vercel backend, and set your own weights and thresholds. Hosted installs are capped at 250 checks per day with no account.

What does each post get scored on?

Twelve questions asked in parallel: nine signs of slop taken from Graphite's AI tells research, two counter-signs asking whether it sounds like one particular person and whether it is useful to a reader, and one that asks how likely a model drafted the post, which can only soften the score and never flags a post by itself.

What are the hide and highlight modes?

Hide collapses suspected slop into a compact strip you can reopen, while Highlight leaves the post in place, outlines it yellow or red and explains why. Mild, Moderate and Aggressive set how much doubt it takes to act.

How do votes and draft checks work?

Every post gets a mop icon; open it to see the score out of 100, a spider chart of what Jev noticed, and to vote No, Maybe or Probably, which overrides the model for you. The composer also gets a mop button that scores your draft as your own writing using one of the day's checks, and nothing is read until you press it.

CommunityJev Tools & CLIs

Slopmop

Chromium extension plus self-hostable server that scores LinkedIn post text with twelve parallel Jev writing judgments, then folds or outlines suspected slop with reasons you can overrule.

See highlights

Open github.com

Stars: 3
Language: TypeScript
License: MIT
Published by: Community
Author: tomfrazier
Last verified: Oct 2026

tomfrazier/slopmop

User repository on GitHub

View on GitHub

Mop the slop out of your LinkedIn feed. Not an AI detector: a bad-writing detector. Chrome extension + Vercel server, powered by Jev from Typesafe AI. MIT.

Forks: 1
Last pushed: Updated Oct 2026

Repo stats from the GitHub API, cached Oct 2026.Homepage

About this project

Mop the slop out of your LinkedIn feed. Not an AI detector. A bad-writing detector. Your attention deserves a filter you control. Free. No signup. MIT licensed. Desktop Chrome and other Chromium browsers, LinkedIn only for now.

Quoted from the project's README. Read the full README.

Tagscommunity github

Works forJev for Classification

Used inJev for Software & DevTools

Languages: TypeScript

Highlights

Asks nine slop signs from Graphite tells research plus two counter-signs for voice and usefulness.
Uses AI-drafted likelihood only to soften human-seeming scores, never as a standalone flag.
Offers Hide, Highlight, and Mild, Moderate, and Aggressive thresholds with per-post reasons.
Sends post text for scoring, stores only hashes and tallies, and skips names, drafts, ads, and tabs.

Watch out

LinkedIn and desktop Chromium only; the repo ships equal weights while hosted scores use tuned weights, fitted on 27 hand votes. Copy reviewed against the repository README on 2026-10-04, not from running it.

Questions about this entry

Short answers drawn from the linked project or page.

More Jev resources like this

Jevest

GitHub Action that reviews pull requests with a read-only finder agent plus a refuting verifier, using Jev only to triage, route, label, and gate merging.