User repository on GitHub
Jevによるnote記事のAI Slop判定サイト
- Forks
- 0
- Last pushed
- Updated Sep 2026
Repo stats from the GitHub API, cached Oct 2026.Homepage
CommunityJev Tools & CLIs
JevSlop
A bring-your-own-key tool that pastes a public note.com article URL and returns Jev's whole-article writing-quality judgment plus eight detail signals as an AI Slop Score. Cloudflare Pages deploy, MIT licensed, explicitly not an authorship detector.
- Stars
- 4
- Language
- TypeScript
- License
- MIT
- Published by
- Community
- Author
- TKY-27
- Last verified
- Oct 2026
About this project
JevSlop asks TypeSafe Jev for a whole-article writing-quality judgment and eight detail signals. It shows a transparent AI Slop Score; it is not an AI-authorship detector. 日本語版. Paste a public /n/ URL, then choose Analyze. JevSlop fetches the article, extracts its title and visible body, asks Jev for a whole-article AI Slop judgment and score, and shows eight separate writing-quality detail signals. Japanese and English UI, selected from the browser language on first visit.
Quoted from the project's README. Read the full README.
Used inJev for Software & DevTools
- Languages
- TypeScript
Highlights
- Paste a public https://note.com/<user>/n/<article-id> URL; the app fetches title and visible body, then asks Jev.
- Eight atomic detail Score questions plus an overall Score and choice go out in a single systemOne request.
- The AI Slop Score runs 0-100 with per-axis distributions, confidence, timing and comparison history; exports are CSV and JSON.
- Your own key only: held in memory and sessionStorage, never localStorage, cleared when the tab closes.
Watch out
The README is unusually complete, but the headline number is a model judgment rather than a measurement: the AI Slop Score is Jev's whole-article choice, and the README states plainly that a human article can score high, AI-assisted writing can score low, and the result must not be read as an authorship probability. It also notes confidence is not a correctness guarantee and that Japanese and English calibration can differ. Scope is limited to public, free, full-text note articles — private, paid, deleted, non-HTML, oversized or unsupported-layout pages fail closed — and images, audio and video are never evaluated. Only title and body reach TypeSafe, with URL, author, date and experiment groups excluded, though Cloudflare and TypeSafe still process the request under their own policies. 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.
No, and the README pre-empts this in its first paragraph: it is not an AI-authorship detector. The score describes writing characteristics, so a human article can score high and AI-assisted writing can score low. If you need authorship attribution, this will hand you a confident-looking number that does not mean what you want it to mean.
You bring your own. The Settings dialog behind the fixed gear in the top-right takes a TypeSafe key held only in memory and sessionStorage, never localStorage, cleared when the tab closes, and no operator key is required. The README explicitly warns against putting TYPESAFE_API_KEY into Cloudflare Pages variables for this app. The catch is cost: each analysis is a full-body systemOne call with thinking on, repeated as you iterate.
The browser posts to a same-origin /api/evaluate Pages Function, which validates the note host and redirects, extracts only the title and body, and forwards your key to TypeSafe. TypeSafe receives exactly { title, body } — no URL, author, date or experiment group. The body is not kept in history or exports and history lives in the tab's memory only, though Cloudflare and TypeSafe process it under their own policies.
The deployment is documented down to a settings table: Next.js static export, main branch, npx next build, output directory out, no env vars, and the functions/ directory supplies the Pages Function automatically. Local dev needs Node 22.12+ and npm run dev serves on port 8788. For public operation the README says to configure Cloudflare rate limiting on /api/evaluate, since the Function's Origin and Sec-Fetch-Site checks are not rate limiting or authentication.
It fetches a public note.com article and asks Jev for a whole-article AI Slop judgment and score plus eight separate writing-quality detail signals, all in one systemOne request. The primary AI Slop or Not AI Slop label is Jev's whole-article choice, not an average of the eight axes, and it is not an authorship probability.
TypeSafe receives exactly the title and body. The URL, author, date, comparison labels and experiment groups are excluded from Jev state, and the article body is not kept in browser history or result exports; result history exists only in the current tab's memory. The SDK runs with logging off and retries disabled.
It needs Node.js 22.12+. Run npm install and npm run dev, which builds the static export and starts a Pages Functions-compatible server at http://localhost:8788, then enter your key in Settings. For Cloudflare Pages, connect the repository with the Next.js static HTML export preset, build command npx next build and output directory out, with no environment variables required.
More Jev resources like this
27GitHub stars
Pi extension that replaces many small tool calls with one typed TypeScript execute_tools call, type-checked before running in a sandbox.
Jev Tools & CLIs#community#jevbest#catalog-import
Community
boozedog/pi-codemodeTypeScript
boozedog/pi-codemodeTypeScript17GitHub stars
Pi extension that trims wasted context before each model call by deduplicating repeated tool results, stripping failed payload args, and letting the model summarize closed spans.
Jev Tools & CLIs#community#jevbest#catalog-import
Community
Davidcreador/pi-dcpTypeScript
Davidcreador/pi-dcpTypeScript4GitHub stars
Shell CLI that sends state plus noul, choice, and score questions to Jev in one request and prints probabilities, choices, and rubric positions for pipes and scripts.
Jev Tools & CLIs#community#jevbest#catalog-import
Community
y0usaf/typesafe-cliTypeScript
y0usaf/typesafe-cliTypeScript1GitHub stars
A local React and Express editor that scores an X draft across six dimensions, from hook strength to a slop penalty, by asking Jev several narrow rubric questions in one structured call, with a deterministic fallback when no key is set.
Jev Tools & CLIs#community#jevbest#catalog-import
Community
pekth/draftpulseTypeScript
pekth/draftpulseTypeScript197GitHub stars
A Chinese-first desktop workbench for local models whose JEV page runs typed noul, choice and score questions against a local engine or a cloud key, alongside a gateway, RAG knowledge base and shared memory.
Jev Tools & CLIs#community#jevbest
Community
kunpengtalk/OmniStudioTypeScript
kunpengtalk/OmniStudioTypeScript38GitHub stars
Research dashboard for Polymarket BTC five-minute Up/Down rounds that feeds market state to Jev class scores and paper-trades the edge.
Jev Tools & CLIs#community#jevbest#catalog-import
From the Jev guides
Original write-ups on what this entry is for.
The request shape barely changes across surfaces; the surrounding contract does. When to call the API directly and when a gateway, a Worker binding, or a router entry earns its extra hop.
Jev Tools & CLIs#gateway#vercel#cloudflare
Read article
Taking a decision checkpoint out of Python: what the GGUF and browser WebGPU exports preserve, what they change, and the parity evidence the model cards publish.
Jev Tools & CLIs#open-models#local#onnx
Read article
Jev labels a record that already exists. Code moves, flags, or holds it. How the document, mail, and transfer tools keep the action out of the model.
Jev Tools & CLIs#classification#email#security
Read article
From the Jev community
Posts from builders shipping with Jev right now.