Live Post Judge
Type a post and Jev scores it as you write — hook, evidence, sales intent, engagement bait — re-judged every time you pause. No key.
Reads signals in the wording and returns probabilities — a guess about the text, not a fact about anyone. One Jev call each time you pause.
The same demo is one Jev call
- Send the draft post as you type (debounced)
- Ask Jev to score hook, evidence, sales intent, bait
- One call returns all four
scores - Update the meters live
const res = await fetch("https://jevtypesafeai.com/api/v1/decide", {
method: "POST",
headers: { Authorization: `Bearer ${process.env.JEV_API_KEY}` },
body: JSON.stringify({
state: post, // the draft, as text
questions: {
hook: { type: "score", instructions: "Hook strength?", criteria: ["weak","ok","strong"] },
bait: { type: "score", instructions: "Engagement-bait?", criteria: ["none","some","heavy"] },
},
}),
});
const { answers } = await res.json();
setMeters(answers); // { hook: {score}, bait: {score}, … }See a real decision heavy · 219ms
Draft: “This one trick 10x'd our revenue 🤯 comment ‘GROW’ and I'll DM you the doc”
How it works
Every time you stop typing for about a third of a second, the current text is sent to the Jev API as four typed questions — hook strength, whether it shows evidence, sales intent and engagement bait — and the bars update from the calibrated probabilities that come back, usually in 100–300ms. It's the same call you'd make in your own editor; here it just runs live as you write.
What it reads
Jev reads signals in the wording and returns probabilities — it is a guess about the text, not a fact about the author. Use it to sharpen a hook or catch engagement bait, not to judge anyone.
FAQ
How can it score as I type?
Each pause triggers one Jev API call, and Jev is fast — a typed decision in about 100–300ms — so the scores keep up with your typing.
Is it free? Do I need a key?
Free and no key — rate-limited calls to the live Jev API. To score posts inside your own product, get a hosted key.
Is this a fact-checker?
No. It reads signals in the wording and returns probabilities about the text; it does not verify claims or judge the writer.
More: Viral Post Analyzer · All Jev tools · Jev API docs