shipwithjev

Catalog / Agents & browsers

0249GitHub

x-scanner

Chrome extension that labels every post you scroll past on X with six typed questions per post, and counts what it costs in the corner.

oso95/x-scannerREADME ↗
# x-scanner

Behavioral labels on every post you scroll past on X, judged by [Jev](https://docs.typesafe.ai),
TypeSafe's System One model, with a counter in the corner showing exactly what it cost.

https://github.com/user-attachments/assets/bd9782e9-cd8d-426b-8585-e3473a007d11

Each post is sent to Jev with six typed questions in one request as soon as it comes within 800 px
of the viewport. The answer comes back in about 150 ms as numbers, not prose, and lands in a chip
under the post, usually before you have scrolled to it. Most posts
come back clean. The ones that don't get an orange flag. Scroll for a minute and the panel reads
something like 80 posts, $0.0027.

## Install

Chrome 120 or newer. Until the Chrome Web Store listing is live, install from source; Node 22 or newer
is needed to build.

```sh
git clone https://github.com/oso95/x-scanner.git
cd x-scanner
npm install
npm run build
```

1. Open `chrome://extensions`, turn on Developer mode, click **Load unpacked**, choose the `dist/` folder.
2. Click the x-scanner icon. Paste your TypeSafe API key, click **Test connection**, then **Save**.
3. Open [x.com](https://x.com) and scroll: home, profiles, search, threads, lists.

Your key lives in this browser's extension storage and nowhere else. The only network traffic is the
post text to `api.typesafe.ai`. No server, no analytics.

## What you see

**Under each post**, a chip styled like X's own metadata line:

- `✓ clean` in green when nothing crossed a threshold, followed by every value in gray.
- `⚑ engagement bait 97%` in orange when something did, followed by the rest in gray.
- `no text to analyze` or `promoted, not analyzed` in dashed gray for posts that are skipped.
- Click the chip for a card with a bar per dimension, the token count, cost and latency of

Also filed under Agents & browsers

  1. 0226

    Flight search with Browser Use★

    Breaking: Browser Use + Jev = Ultrafast ⚡ Findings flights took 7s and cost only $0.0039 🤯 > new action space every step > DOM state space > small LLM fallback to type (this video is at 1x speed btw) Built a tiny open source browser agent. try it below ↓

    @gregpr07 · Agents & browsers · ~$0.004 · ~7 s

  2. 0619

    jevmem: automatic project memory for Claude Code

    After every Claude Code message, Jev decides in ~0.3 s whether it holds a decision, rule or bug worth keeping, and jevmem saves it as one line in JEVMEM.md.

    @jetwaniavinash · Agents & browsers · ~0.3 s

  3. 0618

    Playwright + Jev: tests in plain language

    Write Playwright steps like "add the most expensive item to the cart". Jev finds the right buttons and fields, fills forms and checks the result.

    Andrey Popov · Agents & browsers

  4. 0617

    toolgate: a tool-call firewall for agents

    A Claude Code and MCP hook where Jev answers seven yes/no risk questions per tool call to allow, ask or deny. Usage and benchmark numbers are in the repo.

    Jaz · Agents & browsers