Chrome + Firefox extension · MIT · LinkedIn only, for now

Engagement bait, scored out of 10.

chamuy0 reads your LinkedIn feed and puts a chip on every post. Click the chip to see the signals behind the score. Set a threshold and the worst posts get stamped and folded into a thin marker you can reopen.

In review on the Chrome Web Store and addons.mozilla.org. Until the listings are live, the release zip is the only install path. It takes about two minutes.

Two LinkedIn posts side by side: one scored 8.4 slop with its signal breakdown open, one scored 1.6 clean.
The chip, and the breakdown behind it.
  • A chip on every post.

    ✓ 1.6 clean, ≈ 3.4 borderline, ☣ 8.4 slop. Posts shorter than 40 characters are skipped on purpose.

  • Click it for the why.

    The score out of 10, the verdict, and the eight signals sorted by confidence.

  • Train it as you go.

    Label a post clean, borderline or slop. Later classifications use your latest labels as reference.

Six chip states: scoring in progress, clean, borderline, slop, classifier unreachable, and the floating chip on a media post.
Every chip state, including the one for a classifier that did not answer.

How it works

  1. Install. The free trial starts on its own.

    A 1-day trial scores posts with no setup. After it, subscribe in pesos through Mercado Pago or paste your own TypeSafe API key, which stays in your browser profile.

  2. Scroll your feed. Each post gets a score.

    The text of the post goes to api.typesafe.ai to be scored, through our proxy while you are on the trial or a subscription. Nothing else does.

  3. Banish what you do not want.

    Set a threshold. Posts at or above it get stamped and folded away. Nothing is deleted.

A banished post leaves a marker, not a hole.

Posts at or above your threshold collapse into a thin row: the score, the word hidden, and a Show button. Above 8 the stamp gets a ring. The layout does not jump. Every hidden post is one click from coming back.

Jordan Blake

I interviewed a janitor at 5am.

He looked me in the eyes and said: “success is rented, not owned.”

That conversation changed how I lead.

Agree?

Like Comment Repost Slop8.4
Slop 8.4 · hidden

150 ms stamp · 240 ms collapse · cubic-bezier(.2, 0, 0, 1). The same motion the extension runs, at the same speed.

Two frames of the banish motion on the same post: the stamp at 150 milliseconds and the collapsed marker at 600 milliseconds.
The stamp at 150 ms. The marker at 600 ms.

Eight yes/no questions per post.

  • engagement bait
  • rage bait
  • humblebrag
  • AI generic
  • corporate buzzwords
  • broetry
  • fabricated story
  • manufactured hook

Plus one score question with ten levels. The whole rubric is one file: core/rubric.ts.

Free for a day, then $2 a month or your own key.

A 1-day free trial starts on install. After it you can subscribe in pesos (ARS 3.000/month or ARS 15.000/year) through Mercado Pago, no account, or bring your own TypeSafe key and pay TypeSafe directly. Either way, chamuy0 has no account.

$0.042per million input tokens
~230tokens per post
~$0.01per thousand posts

The token cost is what TypeSafe charges your own key. On the trial or a subscription you pay ARS 3.000/month or ARS 15.000/year and nothing per post. Jev is in early access, and long posts cost more than short ones.

The only thing that leaves your browser is the post text.

Stays on your device:

Your TypeSafe key. Your labels and the examples the classifier calibrates with. Your settings, including the hide threshold. All of it in storage.local.

Leaves your device:

The text of the post being scored, sent to api.typesafe.ai. Up to six recent labeled examples ride along as reference. With your own key it goes straight there; on the trial or a subscription it passes through our proxy first.

Never happens:

No account. No analytics. No telemetry. No ads. Nothing posted, commented, or reacted to on your behalf. The proxy keeps only a hashed device fingerprint and a truncated network address, to stop trial restarts, and never stores post text.

Privacy policy

Install from GitHub. About two minutes.

The Chrome Web Store and addons.mozilla.org listings are in review. The release zip is the same build; it just needs a few clicks to load.

Firefox

  1. Download the latest release.
  2. Open about:debugging#/runtime/this-firefox.
  3. Load Temporary Add-on… and pick the zip.
  4. The setup page opens by itself. The trial is already running.
  5. Open linkedin.com.

Firefox drops temporary add-ons when it closes. Repeat steps 2 and 3. The trial does not restart.

Chrome, Edge, Brave

  1. Download and unzip the release.
  2. Open chrome://extensions and turn on Developer mode.
  3. Load unpacked, pick the folder.
  4. Click the extension icon. The trial is already running.
The chamuy0 settings page: API key field, hide threshold slider set to 6.5, and three training examples.
Settings, the first time you open it.

Install asks for three things: storage, linkedin.com, and api.typesafe.ai. Permissions are granted at install; there is no way to ask for less and still read the feed.

FAQ

Does it send my LinkedIn data anywhere?

Only the text of the post being scored, to api.typesafe.ai, through our proxy on the trial or a subscription. Your key and labels stay in your browser profile.

What model scores the posts?

Jev, a decision model from TypeSafe. It returns a score and signals, not text. It cannot write or hallucinate a post.

Do I need my own API key?

No. A free 1-day trial starts on install, and a peso subscription keeps it running without a key. Your own key is the free-forever path: it goes straight from your browser to TypeSafe and never reaches us.

What does it cost to run?

With your own key, about $0.042 per million input tokens, roughly a cent per thousand posts. On the trial or a subscription, ARS 3.000/month or ARS 15.000/year and nothing per post.

Will it hide posts I wanted?

Nothing is deleted. A hidden post becomes a thin marker with a Show button, and you can raise the threshold.

Why is the install manual?

The Chrome Web Store and Firefox listings are in review. The zip is the same build.

Does it work on Firefox?

Yes. Regular Firefox only keeps signed add-ons, so a permanent install means signing the zip on addons.mozilla.org (free, not public) or using Developer Edition. Temporary loading works everywhere.

Does it work anywhere but LinkedIn?

Not yet.

Is it really open source?

MIT. The whole thing, including the rubric, is at github.com/tsu-ld/chamuy0.

Install chamuy0.

MIT. No account. No analytics. LinkedIn only, for now.